array:23 [
  "pii" => "S0210569120301029"
  "issn" => "02105691"
  "doi" => "10.1016/j.medin.2020.04.003"
  "estado" => "S300"
  "fechaPublicacion" => "2022-03-01"
  "aid" => "1486"
  "copyright" => "Elsevier España, S.L.U. y SEMICYUC"
  "copyrightAnyo" => "2020"
  "documento" => "article"
  "crossmark" => 1
  "subdocumento" => "rev"
  "cita" => "Med Intensiva. 2022;46:140-56"
  "abierto" => array:3 [
    "ES" => true
    "ES2" => true
    "LATM" => true
  ]
  "gratuito" => true
  "lecturas" => array:1 [
    "total" => 0
  ]
  "itemSiguiente" => array:19 [
    "pii" => "S0210569121000711"
    "issn" => "02105691"
    "doi" => "10.1016/j.medin.2021.03.012"
    "estado" => "S300"
    "fechaPublicacion" => "2022-03-01"
    "aid" => "1654"
    "copyright" => "Elsevier España, S.L.U. y SEMICYUC"
    "documento" => "article"
    "crossmark" => 1
    "subdocumento" => "sco"
    "cita" => "Med Intensiva. 2022;46:157-60"
    "abierto" => array:3 [
      "ES" => true
      "ES2" => true
      "LATM" => true
    ]
    "gratuito" => true
    "lecturas" => array:1 [
      "total" => 0
    ]
    "es" => array:11 [
      "idiomaDefecto" => true
      "cabecera" => "<span class="elsevierStyleTextfn">Punto de vista</span>"
      "titulo" => "<span class="elsevierStyleItalic">Trigger</span> transfusional en el paciente con traumatismo cr&#225;neo-encef&#225;lico grave"
      "tienePdf" => "es"
      "tieneTextoCompleto" => "es"
      "paginas" => array:1 [
        0 => array:2 [
          "paginaInicial" => "157"
          "paginaFinal" => "160"
        ]
      ]
      "titulosAlternativos" => array:1 [
        "en" => array:1 [
          "titulo" => "Trigger transfusion in severe traumatic brain injury"
        ]
      ]
      "contieneTextoCompleto" => array:1 [
        "es" => true
      ]
      "contienePdf" => array:1 [
        "es" => true
      ]
      "resumenGrafico" => array:2 [
        "original" => 0
        "multimedia" => array:7 [
          "identificador" => "fig0005"
          "etiqueta" => "Figura 1"
          "tipo" => "MULTIMEDIAFIGURA"
          "mostrarFloat" => true
          "mostrarDisplay" => false
          "figura" => array:1 [
            0 => array:4 [
              "imagen" => "gr1.jpeg"
              "Alto" => 1801
              "Ancho" => 3167
              "Tamanyo" => 464570
            ]
          ]
          "descripcion" => array:1 [
            "es" => "<p id="spar0005" class="elsevierStyleSimplePara elsevierViewall">Posibles causas de hipoxia tisular&#46;</p> <p id="spar0010" class="elsevierStyleSimplePara elsevierViewall">CMRO<span class="elsevierStyleInf">2</span>&#58; consumo metab&#243;lico de ox&#237;geno&#59; Hb&#58; hemoglobina&#46;</p>"
          ]
        ]
      ]
      "autores" => array:1 [
        0 => array:2 [
          "autoresLista" => "J&#46;J&#46; Egea-Guerrero, I&#46; Garc&#237;a-S&#225;ez, M&#46; Quintana-D&#237;az"
          "autores" => array:3 [
            0 => array:2 [
              "nombre" => "J&#46;J&#46;"
              "apellidos" => "Egea-Guerrero"
            ]
            1 => array:2 [
              "nombre" => "I&#46;"
              "apellidos" => "Garc&#237;a-S&#225;ez"
            ]
            2 => array:2 [
              "nombre" => "M&#46;"
              "apellidos" => "Quintana-D&#237;az"
            ]
          ]
        ]
      ]
    ]
    "idiomaDefecto" => "es"
    "Traduccion" => array:1 [
      "en" => array:9 [
        "pii" => "S217357272100182X"
        "doi" => "10.1016/j.medine.2021.12.003"
        "estado" => "S300"
        "subdocumento" => ""
        "abierto" => array:3 [
          "ES" => true
          "ES2" => true
          "LATM" => true
        ]
        "gratuito" => true
        "lecturas" => array:1 [
          "total" => 0
        ]
        "idiomaDefecto" => "en"
        "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S217357272100182X?idApp=WMIE"
      ]
    ]
    "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S0210569121000711?idApp=WMIE"
    "url" => "/02105691/0000004600000003/v1_202202250826/S0210569121000711/v1_202202250826/es/main.assets"
  ]
  "itemAnterior" => array:18 [
    "pii" => "S0210569120303363"
    "issn" => "02105691"
    "doi" => "10.1016/j.medin.2020.10.003"
    "estado" => "S300"
    "fechaPublicacion" => "2022-03-01"
    "aid" => "1601"
    "copyright" => "Elsevier Espa&#241;a&#44; S&#46;L&#46;U&#46; y SEMICYUC"
    "documento" => "article"
    "crossmark" => 1
    "subdocumento" => "fla"
    "cita" => "Med Intensiva. 2022;46:132-9"
    "abierto" => array:3 [
      "ES" => true
      "ES2" => true
      "LATM" => true
    ]
    "gratuito" => true
    "lecturas" => array:1 [
      "total" => 0
    ]
    "en" => array:13 [
      "idiomaDefecto" => true
      "cabecera" => "<span class="elsevierStyleTextfn">Original article</span>"
      "titulo" => "High serum nitrates levels in non-survivor COVID-19 patients"
      "tienePdf" => "en"
      "tieneTextoCompleto" => "en"
      "tieneResumen" => array:2 [
        0 => "en"
        1 => "es"
      ]
      "paginas" => array:1 [
        0 => array:2 [
          "paginaInicial" => "132"
          "paginaFinal" => "139"
        ]
      ]
      "titulosAlternativos" => array:1 [
        "es" => array:1 [
          "titulo" => "Altas concentraciones s&#233;ricas de nitratos en pacientes fallecidos por COVID-19"
        ]
      ]
      "contieneResumen" => array:2 [
        "en" => true
        "es" => true
      ]
      "contieneTextoCompleto" => array:1 [
        "en" => true
      ]
      "contienePdf" => array:1 [
        "en" => true
      ]
      "resumenGrafico" => array:2 [
        "original" => 0
        "multimedia" => array:7 [
          "identificador" => "fig0010"
          "etiqueta" => "Figure 2"
          "tipo" => "MULTIMEDIAFIGURA"
          "mostrarFloat" => true
          "mostrarDisplay" => false
          "figura" => array:1 [
            0 => array:4 [
              "imagen" => "gr2.jpeg"
              "Alto" => 1492
              "Ancho" => 1500
              "Tamanyo" => 169009
            ]
          ]
          "descripcion" => array:1 [
            "en" => "<p id="spar0090" class="elsevierStyleSimplePara elsevierViewall">Survival curves at 30 days using serum nitrates concentrations lower or equal vs higher than 68&#46;4<span class="elsevierStyleHsp" style=""></span>&#956;mol&#47;L&#46;</p>"
          ]
        ]
      ]
      "autores" => array:1 [
        0 => array:2 [
          "autoresLista" => "L&#46; Lorente, F&#46; G&#243;mez-Bernal, M&#46;M&#46; Mart&#237;n, J&#46;A&#46; Navarro-Gonz&#225;lvez, M&#46; Argueso, A&#46; Perez, L&#46; Ramos-G&#243;mez, J&#46; Sol&#233;-Viol&#225;n, J&#46;A&#46; Marcos y Ramos, N&#46; Ojeda, A&#46; Jim&#233;nez"
          "autores" => array:12 [
            0 => array:2 [
              "nombre" => "L&#46;"
              "apellidos" => "Lorente"
            ]
            1 => array:2 [
              "nombre" => "F&#46;"
              "apellidos" => "G&#243;mez-Bernal"
            ]
            2 => array:2 [
              "nombre" => "M&#46;M&#46;"
              "apellidos" => "Mart&#237;n"
            ]
            3 => array:2 [
              "nombre" => "J&#46;A&#46;"
              "apellidos" => "Navarro-Gonz&#225;lvez"
            ]
            4 => array:2 [
              "nombre" => "M&#46;"
              "apellidos" => "Argueso"
            ]
            5 => array:2 [
              "nombre" => "A&#46;"
              "apellidos" => "Perez"
            ]
            6 => array:2 [
              "nombre" => "L&#46;"
              "apellidos" => "Ramos-G&#243;mez"
            ]
            7 => array:2 [
              "nombre" => "J&#46;"
              "apellidos" => "Sol&#233;-Viol&#225;n"
            ]
            8 => array:2 [
              "nombre" => "J&#46;A&#46;"
              "apellidos" => "Marcos y Ramos"
            ]
            9 => array:2 [
              "nombre" => "N&#46;"
              "apellidos" => "Ojeda"
            ]
            10 => array:2 [
              "nombre" => "A&#46;"
              "apellidos" => "Jim&#233;nez"
            ]
            11 => array:1 [
              "colaborador" => "Working Group on COVID-19 Canary ICU"
            ]
          ]
        ]
      ]
    ]
    "idiomaDefecto" => "en"
    "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S0210569120303363?idApp=WMIE"
    "url" => "/02105691/0000004600000003/v1_202202250826/S0210569120303363/v1_202202250826/en/main.assets"
  ]
  "en" => array:21 [
    "idiomaDefecto" => true
    "cabecera" => "<span class="elsevierStyleTextfn">Review</span>"
    "titulo" => "Enhancing sepsis management through machine learning techniques&#58; A review"
    "tieneTextoCompleto" => true
    "paginas" => array:1 [
      0 => array:2 [
        "paginaInicial" => "140"
        "paginaFinal" => "156"
      ]
    ]
    "autores" => array:1 [
      0 => array:4 [
        "autoresLista" => "N&#46; Ocampo-Quintero, P&#46; Vidal-Cort&#233;s, L&#46; del R&#237;o Carbajo, F&#46; Fdez-Riverola, M&#46; Reboiro-Jato, D&#46; Glez-Pe&#241;a"
        "autores" => array:6 [
          0 => array:3 [
            "nombre" => "N&#46;"
            "apellidos" => "Ocampo-Quintero"
            "referencia" => array:1 [
              0 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">a</span>"
                "identificador" => "aff0005"
              ]
            ]
          ]
          1 => array:3 [
            "nombre" => "P&#46;"
            "apellidos" => "Vidal-Cort&#233;s"
            "referencia" => array:1 [
              0 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">b</span>"
                "identificador" => "aff0010"
              ]
            ]
          ]
          2 => array:3 [
            "nombre" => "L&#46;"
            "apellidos" => "del R&#237;o Carbajo"
            "referencia" => array:1 [
              0 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">b</span>"
                "identificador" => "aff0010"
              ]
            ]
          ]
          3 => array:3 [
            "nombre" => "F&#46;"
            "apellidos" => "Fdez-Riverola"
            "referencia" => array:3 [
              0 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">a</span>"
                "identificador" => "aff0005"
              ]
              1 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">c</span>"
                "identificador" => "aff0015"
              ]
              2 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">d</span>"
                "identificador" => "aff0020"
              ]
            ]
          ]
          4 => array:3 [
            "nombre" => "M&#46;"
            "apellidos" => "Reboiro-Jato"
            "referencia" => array:3 [
              0 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">a</span>"
                "identificador" => "aff0005"
              ]
              1 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">c</span>"
                "identificador" => "aff0015"
              ]
              2 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">d</span>"
                "identificador" => "aff0020"
              ]
            ]
          ]
          5 => array:4 [
            "nombre" => "D&#46;"
            "apellidos" => "Glez-Pe&#241;a"
            "email" => array:1 [
              0 => "dgpena@uvigo.es"
            ]
            "referencia" => array:4 [
              0 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">a</span>"
                "identificador" => "aff0005"
              ]
              1 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">c</span>"
                "identificador" => "aff0015"
              ]
              2 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">d</span>"
                "identificador" => "aff0020"
              ]
              3 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">&#42;</span>"
                "identificador" => "cor0005"
              ]
            ]
          ]
        ]
        "afiliaciones" => array:4 [
          0 => array:3 [
            "entidad" => "ESEI &#8211; Escuela Superior de Ingenier&#237;a Inform&#225;tica&#44; Universidad de Vigo&#44; Ourense&#44; Spain"
            "etiqueta" => "a"
            "identificador" => "aff0005"
          ]
          1 => array:3 [
            "entidad" => "Intensive Care Unit&#44; Complexo Hospitalario Universitario de Ourense&#44; Ourense&#44; Spain"
            "etiqueta" => "b"
            "identificador" => "aff0010"
          ]
          2 => array:3 [
            "entidad" => "CINBIO &#8211; Centro de Investigaciones Biom&#233;dicas&#44; Universidad de Vigo&#44; Vigo&#44; Spain"
            "etiqueta" => "c"
            "identificador" => "aff0015"
          ]
          3 => array:3 [
            "entidad" => "SING Research Group&#44; Galicia Sur Health Research Institute &#40;IIS Galicia Sur&#41;&#44; SERGAS-UVIGO&#44; Spain"
            "etiqueta" => "d"
            "identificador" => "aff0020"
          ]
        ]
        "correspondencia" => array:1 [
          0 => array:3 [
            "identificador" => "cor0005"
            "etiqueta" => "&#8270;"
            "correspondencia" => "Corresponding author&#46;"
          ]
        ]
      ]
    ]
    "titulosAlternativos" => array:1 [
      "es" => array:1 [
        "titulo" => "Mejora en el manejo de sepsis mediante t&#233;cnicas de aprendizaje autom&#225;tico&#58; una revisi&#243;n"
      ]
    ]
    "resumenGrafico" => array:2 [
      "original" => 0
      "multimedia" => array:7 [
        "identificador" => "fig0005"
        "etiqueta" => "Figure 1"
        "tipo" => "MULTIMEDIAFIGURA"
        "mostrarFloat" => true
        "mostrarDisplay" => false
        "figura" => array:1 [
          0 => array:4 [
            "imagen" => "gr1.jpeg"
            "Alto" => 2841
            "Ancho" => 2917
            "Tamanyo" => 399170
          ]
        ]
        "descripcion" => array:1 [
          "en" => "<p id="spar0015" class="elsevierStyleSimplePara elsevierViewall">Graphic representation of simple classifiers from different families&#58; &#40;a&#41; an ANN classifier with one hidden layer&#44; &#40;b&#41; a bi-dimensional SVM&#44; &#40;c&#41; a LR with one input variable&#44; and &#40;d&#41; a simple DT&#46;</p>"
        ]
      ]
    ]
    "textoCompleto" => "<span class="elsevierStyleSections"><span id="sec0005" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0025">Introduction</span><p id="par0005" class="elsevierStylePara elsevierViewall">Sepsis is a major public health problem and a leading cause of death in the world&#46; Although it is not easy to reliably measure incidence and mortality&#44;<a class="elsevierStyleCrossRefs" href="#bib0410"><span class="elsevierStyleSup">1&#8211;4</span></a> most of the recent data shows both an increase in incidence and number of deaths and a decrease in case-fatality&#46;</p><p id="par0010" class="elsevierStylePara elsevierViewall">In addition to the difficulty in obtaining reliable data&#44; the sepsis definition&#44; first established in 1991 &#40;Sepsis-1&#41;&#44; has been updated in 2001 &#40;Sepsis-2&#41; and 2016 &#40;Sepsis-3&#41;&#44;<a class="elsevierStyleCrossRefs" href="#bib0430"><span class="elsevierStyleSup">5&#8211;7</span></a> which makes temporal comparison difficult&#46; A meta-analysis that includes 22 studies published in high-income countries between 1979 and 2015 shows an incidence of 288 hospital-treated sepsis and 148 hospital-treated severe sepsis per 100&#44;000 person-year&#46;<a class="elsevierStyleCrossRef" href="#bib0445"><span class="elsevierStyleSup">8</span></a> When only studies of the last decade are analyzed&#44; incidence increases to 437 hospital-treated sepsis and 270 hospital-treated severe sepsis per 100&#44;000 person-year&#46; Hospital mortality during this period was 17&#37; for sepsis and 26&#37; for severe sepsis&#46;</p><p id="par0015" class="elsevierStylePara elsevierViewall">From a complementary perspective&#44; in Spain the incidence and mortality data are very heterogeneous&#46; Incidence is nearly 100 cases per 100&#44;000 person-year<a class="elsevierStyleCrossRef" href="#bib0450"><span class="elsevierStyleSup">9</span></a> and hospital mortality ranges from 43&#37; to under 20&#37;&#46;<a class="elsevierStyleCrossRefs" href="#bib0450"><span class="elsevierStyleSup">9&#8211;12</span></a></p><p id="par0020" class="elsevierStylePara elsevierViewall">To sum up&#44; although the numbers vary&#44; data seem to confirm that incidence and number of deaths due to sepsis increases but the case-fatality decreases&#46;</p><p id="par0025" class="elsevierStylePara elsevierViewall">In a separate but complementary perspective&#44; it is also necessary to consider the consumption of health resources&#46; In this regard&#44; it has been published that mean cost per severe sepsis episode is around &#36;20&#44;000&#46;<a class="elsevierStyleCrossRefs" href="#bib0470"><span class="elsevierStyleSup">13&#44;14</span></a></p><p id="par0030" class="elsevierStylePara elsevierViewall">Different measures have shown a beneficial impact in terms of reducing mortality&#44; and together with improvement in Intensive Care Unit &#40;ICU&#41; assistance&#44; are the cause of this reduction&#46; To improve survival of septic patients&#44; these measures must be applied as soon as possible&#46; There are three basic mainstays in sepsis management<a class="elsevierStyleCrossRefs" href="#bib0480"><span class="elsevierStyleSup">15&#44;16</span></a>&#58; &#40;i&#41; early administration of adequate antimicrobial therapy&#44;<a class="elsevierStyleCrossRefs" href="#bib0490"><span class="elsevierStyleSup">17&#44;18</span></a> &#40;ii&#41; resuscitation with fluids and vasopressors&#44;<a class="elsevierStyleCrossRef" href="#bib0500"><span class="elsevierStyleSup">19</span></a> and &#40;iii&#41; source control&#46;<a class="elsevierStyleCrossRef" href="#bib0505"><span class="elsevierStyleSup">20</span></a> Sepsis bundles have been the cornerstone of the improvement of the quality of sepsis care since 2005&#46; Nowadays&#44; hour-1 sepsis bundle includes five measures that must be accomplished in the first hour since the suspicion of sepsis&#44;<a class="elsevierStyleCrossRef" href="#bib0510"><span class="elsevierStyleSup">21</span></a> underlining the importance of time in sepsis treatment&#46;</p><p id="par0035" class="elsevierStylePara elsevierViewall">Despite the fact that adherence with management guidelines has been related with mortality reduction in several studies&#44; sepsis bundles compliance is low&#46;<a class="elsevierStyleCrossRefs" href="#bib0515"><span class="elsevierStyleSup">22&#8211;24</span></a> In this context&#44; different approaches have been tried to enhance guidelines compliance&#46; As an example&#44; educational interventions can achieve a temporary improvement in bundles compliance and&#44; even&#44; a reduction in hospital mortality&#44; but this impact is usually transitory&#46;<a class="elsevierStyleCrossRef" href="#bib0530"><span class="elsevierStyleSup">25</span></a></p><p id="par0040" class="elsevierStylePara elsevierViewall">Other types of interventions&#44; such as the design of detection and management programmes for sepsis at the hospital or state level<a class="elsevierStyleCrossRefs" href="#bib0535"><span class="elsevierStyleSup">26&#44;27</span></a> show similar results and seem to be cost-effective&#46;<a class="elsevierStyleCrossRef" href="#bib0470"><span class="elsevierStyleSup">13</span></a></p><p id="par0045" class="elsevierStylePara elsevierViewall">Recently&#44; a new approach to the problem of sepsis has arisen&#44; based on the application of new information technologies&#46; These initiatives range from relatively simple systems of automatic detection of sepsis&#44; using electronic medical record data<a class="elsevierStyleCrossRefs" href="#bib0545"><span class="elsevierStyleSup">28&#44;29</span></a> or computerized protocols&#44;<a class="elsevierStyleCrossRef" href="#bib0555"><span class="elsevierStyleSup">30</span></a> to more sophisticated systems based on Big Data and Artificial Intelligence &#40;AI&#41; designed to detect and even predict sepsis or guide clinical decisions&#46;</p><p id="par0050" class="elsevierStylePara elsevierViewall">Concretely&#44; Machine Learning &#40;ML&#41;&#44; a subfield of AI&#44; has gained attention in the sector of medicine&#46; ML goes beyond classic &#8220;expert systems&#8221;&#44; whose rules are manually coded into them&#44; by creating a new generation of systems built by &#8220;learning&#8221; from big amounts of data and by dealing with a high number variables simultaneously in order to mimic or even improve human clinical decision making&#46;<a class="elsevierStyleCrossRef" href="#bib0560"><span class="elsevierStyleSup">31</span></a> There are applications of ML to almost all medical fields&#46; Some recent reviews of the use of ML in several medical areas have been published&#44; including medical image analysis&#44;<a class="elsevierStyleCrossRef" href="#bib0565"><span class="elsevierStyleSup">32</span></a> cardiovascular medicine&#44;<a class="elsevierStyleCrossRef" href="#bib0570"><span class="elsevierStyleSup">33</span></a> in critical care&#44;<a class="elsevierStyleCrossRef" href="#bib0575"><span class="elsevierStyleSup">34</span></a> or neuro oncology&#46;<a class="elsevierStyleCrossRef" href="#bib0580"><span class="elsevierStyleSup">35</span></a></p></span><span id="sec0010" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0030">Objectives</span><p id="par0055" class="elsevierStylePara elsevierViewall">The main objective of this work is to conduct a narrative review to provide an overview of how specific ML techniques can be used to improve sepsis management&#44; focusing on the following specific tasks&#58; detection&#44; prediction of sepsis&#47;shock&#44; mortality prediction&#44; hospital stay&#44; costs and adherence to guides&#46; Non-AI or management systems for sepsis treatment&#44; as well as non ML-based expert systems&#44; fall outside the scope of this paper and have not being included in the bibliographic research done&#46;</p><p id="par0060" class="elsevierStylePara elsevierViewall">This review has been designed with the aim of being useful&#44; mainly&#44; for clinicians wanting to know how ML could help them in their daily practice&#44; but also for researchers conducting their own studies applying ML to any sepsis treatment related task&#46; Keeping this in mind&#44; five main questions of interest for clinicians have been defined during the design phase and answered in this paper after a broad bibliographic research&#46; For those readers with no background in ML&#44; a brief explanation of the main ML techniques employed in the papers found during the bibliographic research was included&#46; In addition&#44; the technical information of the studies covered in this review&#44; such as the frequency at which each ML technique was used to solve different tasks&#44; the size and design of the studies or the performance results&#44; may help researchers to design their own approaches in the field&#46;</p></span><span id="sec0015" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0035">Methodology</span><p id="par0065" class="elsevierStylePara elsevierViewall">We conducted broad queries through different search engines&#44; including PubMed&#44; Google Scholar&#44; ResearchGate&#44; and ScienceDirect&#44; using the following keywords&#58; sepsis&#44; machine learning&#44; early&#44; prediction&#44; severe sepsis&#44; mortality&#44; detection&#44; artificial intelligence&#44; and data mining&#46; The last search was conducted on November 7&#44; 2019&#46;</p><p id="par0070" class="elsevierStylePara elsevierViewall">Titles and abstracts of the initially gathered contributions were checked to exclude those papers that fall outside the scope of this study&#44; such as studies related to neonatal sepsis&#46; After that&#44; 34 papers remained from international conferences and renowned journals published in the last 13 years &#40;2007-November 2019&#41;&#46; From the careful reading of the selected contributions&#44; three main tasks were initially identified as the fundamental objective related to sepsis management&#58; &#40;i&#41; sepsis detection&#44; focused on the identification of septic patients&#44; &#40;ii&#41; sepsis prediction&#44; focused on determining which patients are in risk of developing sepsis&#44; and &#40;iii&#41; mortality prediction&#44; focused on determining which septic patients are in risk of death&#46; <a class="elsevierStyleCrossRef" href="#tbl0005">Table 1</a> summarizes the type techniques proposed in the reviewed papers for dealing with these tasks&#46;</p><elsevierMultimedia ident="tbl0005"></elsevierMultimedia><p id="par0075" class="elsevierStylePara elsevierViewall"><a class="elsevierStyleCrossRef" href="#sec0075">Supplementary material S1</a> provides a detailed view of the main features that characterize all the included studies&#44; while <a class="elsevierStyleCrossRefs" href="#tbl0010">Tables 2&#8211;4</a> contain information about the ML techniques and features used on each study and the performance results achieved for the different tasks&#46;</p><elsevierMultimedia ident="tbl0010"></elsevierMultimedia><elsevierMultimedia ident="tbl0015"></elsevierMultimedia><elsevierMultimedia ident="tbl0020"></elsevierMultimedia><p id="par0080" class="elsevierStylePara elsevierViewall">Based on the study of the aforementioned papers&#44; the following section introduces the main ML concepts and algorithms used in these papers&#46; This section is followed by five sections that try to answer key questions specifically related to actual challenges present in sepsis management&#46; The information provided is of special relevance to researchers who want to contribute advances in the area&#46;</p></span><span id="sec0020" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0040">Machine learning overview</span><p id="par0085" class="elsevierStylePara elsevierViewall">ML is a discipline of the AI field focused on making machines able to do tasks without being explicitly programmed for them&#46; To do so&#44; algorithms need to be trained&#44; which&#44; depending on the algorithm&#44; can be done by analyzing sample&#44; or training&#44; data or by iteratively developing a strategy for solving problems based on rewards or punishments&#46; Among the different types of learning strategies&#44; one of the most widely used is supervised learning&#46; Here&#44; the objective is to create a model able to predict some output value given a set of input variables&#46;</p><p id="par0090" class="elsevierStylePara elsevierViewall">For example&#44; to predict the malignant &#40;vs&#46; benign&#41; condition of a given tissue given some morphological variables&#44; patient data or even the specimen&#39;s image&#46; If the output variable is limited to a known set of values&#44; the task is called classification&#44; whereas if the prediction should be any numerical value within a range&#44; the objective is to do a regression&#46;<a class="elsevierStyleCrossRef" href="#bib0745"><span class="elsevierStyleSup">68</span></a> In supervised learning&#44; algorithms need a set of resolved samples&#44; that is&#44; data including the input variables values along with the correct output&#46; Once trained&#44; these algorithms can make predictions over new samples&#46;</p><p id="par0095" class="elsevierStylePara elsevierViewall">Following&#44; we briefly explain the main supervised ML algorithms used in the studies reviewed in this work&#46;</p><p id="par0100" class="elsevierStylePara elsevierViewall">Artificial Neural Networks &#40;ANN&#41; are inspired in the way that neurons in a human or biological brain work&#46; An ANN is an interconnected group of nodes &#40;artificial neurons&#41; separated into&#44; at least&#44; three layers&#58; &#40;i&#41; input&#44; that receives the sample data&#44; &#40;ii&#41; hidden&#44; that transforms the input values&#44; and &#40;iii&#41; output&#44; that provides the final prediction for each sample&#46; However&#44; it is very common to have more than one single hidden layer&#44; as more layers increase the ability of the ANN to learn more complex problems&#46; <a class="elsevierStyleCrossRef" href="#fig0005">Fig&#46; 1</a>a represents a simple ANN with six input neurons&#44; each one corresponding with a variable&#44; one hidden layer with three neurons&#44; and an output layer with a single neuron that outputs the sepsis probability&#46; As can be seen&#44; nodes on each layer are connected with all the nodes on the next layer&#46; These connections are usually initialized with a random weight&#46; The output layer is typically configured to have a node for each possible output condition&#44; and the activation level of each of these nodes for a sample corresponds with the probability of the sample to belong to the condition associated with the node&#46; During the training&#44; the connection weights are adjusted&#44; so that input values can be transformed into the expected output by successively applying mathematical transformations based on the connection weights&#46; ANN classifiers generally have good performance&#44; however&#44; they are complex to configure&#44; costly to train&#44; and difficult to interpret&#46;</p><elsevierMultimedia ident="fig0005"></elsevierMultimedia><p id="par0105" class="elsevierStylePara elsevierViewall">Support Vector Machines &#40;SVM&#41; treat each sample as a point in a n-dimensional space and try to find a hyperplane that separates samples from two different conditions&#46; For this reason&#44; SVM are considered binary classifiers&#44; as they can only classify data with two possible outputs&#46; However&#44; there are several strategies to use SVM in problems with more than two possible outputs by combining several SVM classifiers&#46; <a class="elsevierStyleCrossRef" href="#fig0005">Fig&#46; 1</a>b represents a simple SVM with two input variables &#40;i&#46;e&#46; temperature and heart rate&#41;&#46; During the training&#44; hyperplane parameters are adjusted to maximize the margin between the hyperplane and the samples with different outputs&#46; SVM classifiers generally have good performance&#44; however&#44; they are complex to configure and trained classifiers are difficult to interpret&#46;</p><p id="par0110" class="elsevierStylePara elsevierViewall">Logistic Regression &#40;LR&#41; takes its name from the logistic function on which it is based&#46; Like SVM&#44; LR algorithms are considered binary classifiers&#44; but&#44; in this case&#44; the output will be close or equal to 0 for one condition and close or equal to 1 for the other condition&#46; <a class="elsevierStyleCrossRef" href="#fig0005">Fig&#46; 1</a>c represents a simple LR with a single input variable &#40;i&#46;e&#46; heart rate&#41;&#46; In this example&#44; the sepsis probability increases with the heart rate&#44; however this variable is not enough to predict sepsis&#44; and some samples are misclassified&#46; There is a generalized LR for multiple condition problems named Multinomial Logistic Regression&#46; During the training process&#44; the coefficients of the function used to generate the output are adjusted&#44; so that the output value is as close as possible to 0 or 1 for each condition&#46; Despite of its name&#44; LR is used in classification problems&#46; LR classifiers perform well on linear separable problems and are simple to configure and train&#46; However&#44; trained classifiers are difficult to interpret&#46;</p><p id="par0115" class="elsevierStylePara elsevierViewall">A Decision Tree &#40;DT&#41; algorithm has the internal structure of a tree graph&#44; where each branch node evaluates a variable of the sample&#44; each edge represents a possible outcome of the evaluation and each leaf node represents a possible outcome&#46; <a class="elsevierStyleCrossRef" href="#fig0005">Fig&#46; 1</a>d represents a simple DT that determines if a patient is septic following the sepsis-3 definition&#46; The structure of the tree&#44; including the number of nodes&#44; the condition evaluated on each node and the value of each edge&#44; is determined during the training process&#46; A special type of DT is the Random Forest &#40;RF&#41;&#44; which is not strictly a tree but a set of DTs that use different variables of the samples&#46; The final outcome of a RF is an average of the internal trees outputs&#44; usually the mode&#44; for classification&#44; and the mean&#44; for regression&#46; RF classifiers generally have high performance&#44; they are easy to configure and it is possible to get some useful information to interpret trained classifiers&#44; such as the importance of the variables&#46;</p><p id="par0120" class="elsevierStylePara elsevierViewall">Bayesian algorithms&#44; such as Na&#239;ve Bayes &#40;NB&#41; or Bayesian Network &#40;BN&#41;&#44; are based on Bayes&#8217; theorem&#44; which describes the probability of an event&#44; based on prior knowledge of conditions that might be related to the event&#46; This kind of algorithms use the training data to estimate the probabilities of each possible output based on the values of the variables of the training samples&#46; NB assumes variable independence&#44; which is not required by the BN&#46; Bayesian algorithms are mainly used in text-based problems&#44; although they can be used in many other domains&#46; They are easy to configure and the trained classifiers are interpretable&#46;</p></span><span id="sec0025" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0045">What are the key variables to build a clinical decision support system to assist in sepsis management&#63;</span><p id="par0125" class="elsevierStylePara elsevierViewall">Taking into consideration the conducted review&#44; and focusing the attention on the three main tasks identified in <a class="elsevierStyleCrossRef" href="#tbl0005">Table 1</a> &#40;i&#46;e&#46; sepsis detection&#44; sepsis prediction&#44; and mortality prediction&#41;&#44; we can see there are four different but complementary sources of information that are of utmost importance&#46; First of all&#44; a substantial number of papers mention vital signs &#40;e&#46;g&#46; as heart rate &#40;HR&#41;&#44; respiratory rates &#40;RR&#41;&#44; temperature &#40;Temp&#41;&#44; blood pressure &#40;BP&#41;&#44; oxygen saturation &#40;SaO<span class="elsevierStyleInf">2</span>&#41;&#44; etc&#46;&#41;&#44; and laboratory tests &#40;e&#46;g&#46; renal and liver function&#44; lactate level&#44; coagulation profile&#44; etc&#46;&#41;&#46; Additionally&#44; certain patient characteristics &#40;e&#46;g&#46; age&#44; nationality&#44; comorbidities&#41; are taken into consideration to hypothesize different scenarios&#44; whilst different severity scores such as SOFA &#40;Sequential Organ Failure Assessment&#41;&#44; APACHE &#40;Acute Physiology and Chronic Health Evaluation&#41;&#44; qSOFA &#40;Quick Sequential Organ Failure Assessment&#41;&#44; or MEWS &#40;Modified Early Warning Scoring&#41; are also used by multiple systems&#46; Furthermore&#44; some works emphasize the need to analyze complementary sources of information containing unstructured text such as nursing and medical notes&#44; comments from different departments and personal of emergency with the goal of identifying clues about patients with sepsis&#46;<a class="elsevierStyleCrossRef" href="#bib0645"><span class="elsevierStyleSup">48</span></a></p><p id="par0130" class="elsevierStylePara elsevierViewall">From a complementary perspective&#44; some works indicate that a better management of clinical and administrative databases containing supplementary patient information is necessary to facilitate an automated prediction&#46;<a class="elsevierStyleCrossRefs" href="#bib0595"><span class="elsevierStyleSup">38&#44;46&#44;58&#44;63</span></a> In this line&#44; the Electronic Health Records &#40;EHR&#41;&#44; already available in some centres and ICUs&#44; eases the compilation&#44; interchange&#44; comparison&#44; and effective use of medical information between different departments&#46;</p></span><span id="sec0030" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0050">Can sepsis be automatically detected earlier through the use of ML techniques&#63;</span><p id="par0135" class="elsevierStylePara elsevierViewall">Sepsis is a time-dependent syndrome&#44; which implies that the sooner we start treatment&#44; the better the prognosis&#46;<a class="elsevierStyleCrossRefs" href="#bib0480"><span class="elsevierStyleSup">15&#44;18</span></a> Although many efforts have been carried out in order to establish and update a precise definition of sepsis&#44;<a class="elsevierStyleCrossRefs" href="#bib0430"><span class="elsevierStyleSup">5&#8211;7</span></a> the current identification of septic patients remains troublesome&#46; An accurate sepsis detection system may change the way we approach its treatment&#46;</p><p id="par0140" class="elsevierStylePara elsevierViewall">Of the 34 works covered in the present study&#44; 15 of them explicitly deal with the task of automatically detect sepsis&#46;<a class="elsevierStyleCrossRefs" href="#bib0585"><span class="elsevierStyleSup">36&#8211;42&#44;46&#8211;50</span></a> Main features of the most relevant works are commented hereunder&#46;</p><p id="par0145" class="elsevierStylePara elsevierViewall">Mao et al&#46;<a class="elsevierStyleCrossRef" href="#bib0615"><span class="elsevierStyleSup">42</span></a> developed a ML-based system &#40;called InSight&#41; that recognizes sepsis using only six common vital signs taken from the EHR of Emergency Department &#40;ED&#41;&#44; ICU or hospital wards patients &#40;SaO<span class="elsevierStyleInf">2</span>&#44; HR&#44; diastolic blood pressure &#40;DBP&#41;&#44; systolic blood pressure &#40;SBP&#41;&#44; Temp&#44; and RR&#41;&#46; The authors reported that InSight works correctly despite a significant amount of missing patient data&#46; Faisal et al&#46;<a class="elsevierStyleCrossRef" href="#bib0640"><span class="elsevierStyleSup">47</span></a> developed a LR model to predict the risk of sepsis using only the first electronically recorded vital signs and adding first blood test results obtained following emergency medical admission&#44; with good performance&#46; Horng et al&#46;<a class="elsevierStyleCrossRef" href="#bib0645"><span class="elsevierStyleSup">48</span></a> demonstrated that the use of free text&#44; in addition to vital signs and demographic information&#44; allows for an increase in the ability of identifying infection at ED triage&#46; Delahanty et al&#46;<a class="elsevierStyleCrossRef" href="#bib0620"><span class="elsevierStyleSup">43</span></a> and Barton et al&#46;<a class="elsevierStyleCrossRef" href="#bib0630"><span class="elsevierStyleSup">45</span></a> showed us that ML based models have a better performance than usually used &#40;and recommended by clinical guidelines&#41; scores&#44; like qSOFA&#44; NEWS or MEWS&#46;</p><p id="par0150" class="elsevierStylePara elsevierViewall">As we can see&#44; most of ML-based sepsis detection systems use age&#44; well-known vital signs &#40;i&#46;e&#46; HR&#44; RR&#44; Temp&#44; SBP&#44; and DBP&#41; and usual blood test results to build their models&#46; Other authors use less common parameters&#46; Arvind et al&#46;<a class="elsevierStyleCrossRef" href="#bib0655"><span class="elsevierStyleSup">50</span></a> showed us an interesting way of detecting sepsis using solely unstructured narrative discharge notes &#40;Area Under Curve &#40;AUC&#41; 0&#46;82&#41;&#46; The work of Tang et al&#46;<a class="elsevierStyleCrossRef" href="#bib0650"><span class="elsevierStyleSup">49</span></a> is particular because of their use of electrocardiogram and finger and ear-lobe photoplethysmography signals to detect sepsis &#40;AUC 0&#46;78 for severe sepsis&#41;&#46; There are&#44; also&#44; more experimental strategies&#44; such as the one proposed by Wang et al&#46;&#44;<a class="elsevierStyleCrossRef" href="#bib0585"><span class="elsevierStyleSup">36</span></a> who employed five biomarkers &#40;<span class="elsevierStyleSmallCaps">d</span>-xylose&#44; Acetatic acid&#44; Linoleic acid&#44; <span class="elsevierStyleSmallCaps">d</span>-glucopyranosiduronic acid&#44; and cholesterol&#41;&#44; reaching a sensibility of 0&#46;896 and a specificity of 0&#46;658&#46;</p><p id="par0155" class="elsevierStylePara elsevierViewall">To sum up&#44; ML based model have a good performance to detect sepsis&#44; even with a few vital signs obtained routinely&#44; and without need for suspicion of sepsis&#44; at ED admission or during hospital stay&#46; Specific data and results of sepsis detection studies in <a class="elsevierStyleCrossRef" href="#tbl0010">Table 2</a>&#46;</p></span><span id="sec0035" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0055">Is sepsis prediction an achievable goal&#63;</span><p id="par0160" class="elsevierStylePara elsevierViewall">As we have already explained&#44; sepsis is a time-dependent syndrome&#44; so we must detect it as fast as we can&#46; Would it be possible even to predict that an infected patient is going to have a sepsis a few hours before its appearance or that a sepsis is going to deteriorate into septic shock&#63; There are some approaches that expressly address this challenge&#46;</p><p id="par0165" class="elsevierStylePara elsevierViewall">Of the 34 works covered in the present study&#44; 12 of them explicitly deal with the task of sepsis prediction and sepsis severity prediction&#46; From a quantitative perspective&#44; some studies were carried out using large datasets&#46;<a class="elsevierStyleCrossRefs" href="#bib0600"><span class="elsevierStyleSup">39&#44;41&#44;50&#44;58</span></a></p><p id="par0170" class="elsevierStylePara elsevierViewall">Mao et al&#46;<a class="elsevierStyleCrossRef" href="#bib0615"><span class="elsevierStyleSup">42</span></a> built a model using only six vital signs able to predict evolution &#40;4<span class="elsevierStyleHsp" style=""></span>h before its appearance&#41; to severe sepsis &#40;AUC 0&#46;85&#41;&#44; and to septic shock &#40;AUC 0&#46;96&#41; in septic patients in ED&#44; hospital wards&#44; and ICU&#46; Using vital signs and combining them with other variables &#40;laboratory results&#44; treatment received&#44; etc&#46;&#41;&#44; Lin et al&#46;<a class="elsevierStyleCrossRef" href="#bib0750"><span class="elsevierStyleSup">69</span></a> and Liu et al&#46;<a class="elsevierStyleCrossRef" href="#bib0755"><span class="elsevierStyleSup">70</span></a> achieved a higher prediction capacity&#44; being able to predict evolution to severe sepsis &#40;AUC 0&#46;94&#41; or septic shock &#40;AUC 0&#46;82&#41; before severe sepsis &#40;12<span class="elsevierStyleHsp" style=""></span>h before&#41; and septic shock &#40;9<span class="elsevierStyleHsp" style=""></span>h before&#41; appearance&#46;</p><p id="par0175" class="elsevierStylePara elsevierViewall">It seems more interesting and challenging to predicting those non-septic patients admitted to hospital wards or ICU who will suffer from sepsis&#46; Some authors have developed systems able to predict sepsis using just a few variables that are usually collected by the EHR &#40;such as age&#44; HR&#44; RR&#44; BP&#44; and SaO<span class="elsevierStyleInf">2</span>&#41;&#46; Scherpf et al&#46;&#44;<a class="elsevierStyleCrossRef" href="#bib0665"><span class="elsevierStyleSup">52</span></a> combining 10 variables &#40;age&#44; DBP&#44; SBP&#44; pH&#44; SaO<span class="elsevierStyleInf">2</span>&#44; Temp&#44; HR&#44; RR&#44; partial pressure of carbon dioxide &#40;PaCO<span class="elsevierStyleInf">2</span>&#41;&#44; and white blood cell &#40;WBC&#41;&#41; achieved an AUC of 0&#46;81 to predict sepsis in ICU patients 3<span class="elsevierStyleHsp" style=""></span>h before its appearance&#46; Barton et al&#46;&#44;<a class="elsevierStyleCrossRef" href="#bib0630"><span class="elsevierStyleSup">45</span></a> using just six variables &#40;SaO<span class="elsevierStyleInf">2</span>&#44; HR&#44; SBP&#44; DBP&#44; Temp&#44; and RR&#41;&#44; were able to predict sepsis development in hospital&#44; ICU&#44; and ED patients&#44; with an advance of 24<span class="elsevierStyleHsp" style=""></span>h &#40;AUC 0&#46;84&#41; and 48<span class="elsevierStyleHsp" style=""></span>hours &#40;AUC 0&#46;83&#41;&#46;</p><p id="par0180" class="elsevierStylePara elsevierViewall">Other more complex approaches&#44; as Giannini&#39;s et al&#46;&#44;<a class="elsevierStyleCrossRef" href="#bib0685"><span class="elsevierStyleSup">56</span></a> combined up to 587 variables to achieve an AUC of 0&#46;88 to predict sepsis in hospital wards patients with one hour advance&#46; As we can see&#44; there are different systems that have demonstrated their ability to predict sepsis before its appearance&#44; some of them based on a few variables usually recorded on EHR&#44; with a good performance&#46; Of course&#44; accuracy increases closer to sepsis onset and when the system can have more dates of trends of each variable&#46; Specific data and results of sepsis prediction in <a class="elsevierStyleCrossRef" href="#tbl0015">Table 3</a>&#46;</p></span><span id="sec0040" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0060">Can ML techniques be used to accurately predict sepsis-related mortality&#63;</span><p id="par0185" class="elsevierStylePara elsevierViewall">Predicting sepsis-related mortality is important in order to both classify patients by their severity and identify those situations in which a more aggressive treatment may be necessary&#46; Moreover&#44; it could be useful for designing clinical trials&#44; aiding in the selection of the target population&#46; A tool that makes mortality prediction would be very useful for the selection of patients to be included in these studies&#46;</p><p id="par0190" class="elsevierStylePara elsevierViewall">Of the 34 works covered in the present study&#44; 9 of them explicitly deal with the task of predicting sepsis-related mortality&#46; There is a significant variability in the size of patient databases&#44; ranging from large datasets to a few individuals under consideration&#46;<a class="elsevierStyleCrossRef" href="#bib0700"><span class="elsevierStyleSup">59</span></a> In most cases&#44; prediction studies were carried out using smaller datasets when compared with the previous task&#46;<a class="elsevierStyleCrossRefs" href="#bib0700"><span class="elsevierStyleSup">59&#44;62</span></a></p><p id="par0195" class="elsevierStylePara elsevierViewall">Both the ability to predict short-term and long-term mortality have been analyzed by different authors&#46;</p><p id="par0200" class="elsevierStylePara elsevierViewall">Short-term mortality prediction has been studied by Perng et al&#46;<a class="elsevierStyleCrossRef" href="#bib0705"><span class="elsevierStyleSup">60</span></a> They used 53 clinical variables&#44; all of them obtained during ED patient stay&#44; and achieved an AUC of 0&#46;94 to predict mortality at 72<span class="elsevierStyleHsp" style=""></span>h and of 0&#46;92 at 28 days&#46; Similar results &#40;but no better&#41; showed Taylor et al&#46;<a class="elsevierStyleCrossRef" href="#bib0720"><span class="elsevierStyleSup">63</span></a> with a more complex algorithm that combines more than 500 variables and achieves an AUC of 0&#46;86 to predict sepsis related mortality at 28 days in ED patients&#46;</p><p id="par0205" class="elsevierStylePara elsevierViewall">Garcia-Gallo et al&#46;<a class="elsevierStyleCrossRef" href="#bib0740"><span class="elsevierStyleSup">67</span></a> developed a ML-based model for predicting 1-year mortality in critically ill patients diagnosed with sepsis&#44; using the MIMIC-III critical care database&#46;<a class="elsevierStyleCrossRef" href="#bib0760"><span class="elsevierStyleSup">71</span></a> Reported results using a tree-based ensemble classifier outperformed those obtained by other traditional scoring systems &#40;e&#46;g&#46; SAPS II&#44; SOFA or OASIS&#41;&#44; reaching an AUC of 0&#46;804&#46; In their model&#44; they include 47 variables &#40;including vital signs&#44; laboratory results&#44; comorbidities&#44; and organ dysfunction&#41;&#46;</p><p id="par0210" class="elsevierStylePara elsevierViewall">Regarding the variables used&#44; most of the proposed models use age&#44; lactate&#44; WBC count&#44; and other well-known vital signs &#40;as RR&#44; Temp&#44; and mean blood pressure&#41; all of them easily accessible in different settings&#46; From a complementary point of view&#44; the work of Byrne et al&#46;<a class="elsevierStyleCrossRef" href="#bib0700"><span class="elsevierStyleSup">59</span></a> is especially interesting because mortality is predicted by using 939 peptides identified with LC-MS&#47;MS &#40;Liquid Chromatography with tandem Mass Spectrometry&#41; in blood samples&#46;</p><p id="par0215" class="elsevierStylePara elsevierViewall">To sum up&#44; different works have shown a relatively good performance predicting long and short-term sepsis related mortality&#44; usually employing more variables than in previous tasks&#46; Specific data and results of mortality prediction in <a class="elsevierStyleCrossRef" href="#tbl0020">Table 4</a>&#46;</p></span><span id="sec0045" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0065">Is it possible to effectively increase adherence to treatment guidelines and reduce sepsis-related mortality and&#47;or associated costs using ML techniques&#63;</span><p id="par0220" class="elsevierStylePara elsevierViewall">Despite the fact that it is difficult to assess how sepsis detection &#40;and especially&#44; sepsis prediction&#41; will change sepsis treatment&#44; in this work we have tried to elucidate if ML techniques could improve&#44; by themselves&#44; sepsis treatment&#46; Due to the high mortality related to sepsis&#44;<a class="elsevierStyleCrossRefs" href="#bib0445"><span class="elsevierStyleSup">8&#44;10</span></a> increasing survival should be the goal when designing an intervention&#46; However&#44; sepsis also generates huge health resources consumption&#44;<a class="elsevierStyleCrossRefs" href="#bib0470"><span class="elsevierStyleSup">13&#44;14</span></a> so a parallel objective should be to reduce the cost per episode&#46; In our review&#44; we found that both objectives &#40;i&#46;e&#46; clinical and economic&#41; could be achieved through the effective use of ML techniques&#46;</p><p id="par0225" class="elsevierStylePara elsevierViewall">The AI Clinician&#44;<a class="elsevierStyleCrossRef" href="#bib0765"><span class="elsevierStyleSup">72</span></a> a computational model using reinforcement learning&#44; developed from the analysis of treatment received by patients in the MIMIC-III&#44; is able to dynamically suggest optimal treatments for adult patients with sepsis&#46; The system has been able to identify optimal fluid and vasopressors management from suboptimal training examples&#46; In an independent validation cohort&#44; comprising patients from the eICU Collaborative Research Database&#44;<a class="elsevierStyleCrossRef" href="#bib0770"><span class="elsevierStyleSup">73</span></a> those who received the treatment that the AI Clinician would recommend had the lowest mortality rate&#46; Authors suggest that this system could be used in a real environment&#44; proposing a course of action for the septic patients in real-time&#46; Although they do not expect to replace the physician&#44; as the selection of the treatment strategy still would require their clinical judgement&#44; they think that the system could provide additional insight about optimal decisions to increase the patient survival expectative&#46;</p><p id="par0230" class="elsevierStylePara elsevierViewall">There are&#44; also&#44; systems that are prospectively tested&#46; The impact of the InSight system&#44;<a class="elsevierStyleCrossRef" href="#bib0615"><span class="elsevierStyleSup">42</span></a> described above&#44; was assessed in two prospective studies&#46; McCoy and Das<a class="elsevierStyleCrossRef" href="#bib0775"><span class="elsevierStyleSup">74</span></a> showed how after deploying InSight &#40;compared with pre-implantation period&#41;&#44; in-hospital mortality rate decreased by 60&#46;24&#37;&#44; sepsis-related hospital length of stay &#40;LOS&#41; decreased by 9&#46;55&#37;&#44; and sepsis-related 30-day readmission rate decreased by 50&#46;14&#37;&#46; The study was carried out with 1328 cases and showed how early intervention can reduce mortality and LOS&#44; thereby decreasing the overall hospital cost&#46; Furthermore&#44; Shimabukuro et al&#46;<a class="elsevierStyleCrossRef" href="#bib0780"><span class="elsevierStyleSup">75</span></a> conducted a randomized clinical trial in two ICUs with the goal of calculating the average LOS and in-hospital mortality rate of two groups of patients &#40;i&#46;e&#46; 67 experimental patients vs&#46; 75 control patients&#41;&#46; InSight obtained a decrease of the average LOS of 2&#46;7 days between control and the experimental group &#40;representing a 20&#46;6&#37; reduction&#41;&#46; Additionally&#44; the mortality rate showed a decrease of 58&#46;0&#37;&#46;</p><p id="par0235" class="elsevierStylePara elsevierViewall">Giannini et al&#46;<a class="elsevierStyleCrossRef" href="#bib0685"><span class="elsevierStyleSup">56</span></a> also tested their system in a pre-post trial&#44; resulting just in an increase of lactate testing and intravenous fluids administration with a reduction in time to ICU-admission without impact on mortality or ICU LOS before ML system was implanted&#46;</p></span><span id="sec0050" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0070">Discussion and conclusions</span><p id="par0240" class="elsevierStylePara elsevierViewall">In the future&#44; application of ML systems to sepsis diagnosis and management could change our way of dealing with this pathology&#46; First of all&#44; criteria for sepsis diagnosis could change&#58; until now&#44; the simplicity of the definition was a priority&#44; since sepsis can occur at any level of care and must be recognized by professionals not specialized in this field&#46; This simplification has led us to the point that&#44; nowadays&#44; sepsis diagnosis is based on the presence of established organic dysfunction&#46; A sepsis diagnosis tool based on ML could analyze a massive number of variables not affordable to us&#46; This system&#44; properly calibrated&#44; would allow us to reach a more precise diagnosis&#44; and even predict the appearance of sepsis&#44; completely changing the management of this entity&#46; Secondly&#44; the development of clinical decision support systems &#40;CDSS&#41; with complex algorithms based on AI may improve adherence to accepted management recommendations&#44; giving individualized treatment advice&#46; Finally&#44; employing Big Data and ML based algorithms could let us know a precise outcome for each specific patient&#46;</p><p id="par0245" class="elsevierStylePara elsevierViewall">These systems&#44; although they do not replace human doctors&#44; offer a series of advantages over them&#58; they would allow almost all hospitals around the world to have a sepsis expert present for 24<span class="elsevierStyleHsp" style=""></span>h 7 days a week&#44; who does not get tired&#44; and who always offers evidence-based treatment&#46;</p><p id="par0250" class="elsevierStylePara elsevierViewall">However&#44; ML-based health care is far from ideal&#46; We must also emphasize that there are serious difficulties for the development of intelligent systems for health care&#46; First of all&#44; not all ICUs and hospitals are using EHR today&#44; something that is essential to apply a ML based system&#46; On the other hand&#44; as pointed out by Beam and Kohane&#44;<a class="elsevierStyleCrossRef" href="#bib0785"><span class="elsevierStyleSup">76</span></a> ML &#8220;is not a magic device that can spin data into gold&#8221; directly&#44; as ML is a natural extension of statistics to deal with and take advantage of the huge amounts of data available nowadays&#44; but a big human and scientific effort still is needed to let a machine learn in each specific scenario&#46; In fact&#44; one of the most important elements for ML&#44; if not the most important&#44; is data&#46; ML needs large&#44; experts-curated and&#44; most importantly&#44; labelled&#44; datasets that should be extracted and properly processed&#46; Even though such large datasets could be collected&#44; they may be subject to biases&#46;<a class="elsevierStyleCrossRef" href="#bib0790"><span class="elsevierStyleSup">77</span></a> In this sense&#44; the freely accessible to MIMIC-III critical care database is the most employed source of data for training sepsis-related models&#44; instead of private datasets&#44; showing that it is not easy to find&#46; Although this database has an undoubted value and quality&#44; it only contains ICU patients&#44; where data is recorded very frequently&#46; In fact&#44; many ML efforts go to where data is available&#44; most of times forgetting about its real clinical value&#46;<a class="elsevierStyleCrossRef" href="#bib0795"><span class="elsevierStyleSup">78</span></a> As an example&#44; in the case of sepsis&#44; it is obviously useful to detect or predict sepsis earlier in ICU or ED&#44; using routine variables that are recorded with no need for &#8220;human&#8221; sepsis suspicion&#46; However&#44; what about doing this outside intensive vigilance services&#63; A system predicting this event earlier in hospital wards could be of breakthrough value&#44; but it would require continuous monitoring of the patients with real-time registration of the generated data into their EHR&#44; which is not feasible&#46; On the other hand&#44; despite their high AUC to detect and even predict sepsis&#44; it is very difficult to calibrate these systems&#44; as increasing the specificity and sensitivity of them is usually at the expense of each other&#46; Therefore&#44; if it has a high sensitivity it is going to trigger many unnecessary alarms with the resulting fatigue&#44; whereas if it has a high specificity&#44; some patients are not going to be detected&#46;</p><p id="par0255" class="elsevierStylePara elsevierViewall">One argument against ML could be that some of the relationships stablished by these systems are not explainable from a physiopathological point of view&#44;<a class="elsevierStyleCrossRef" href="#bib0800"><span class="elsevierStyleSup">79</span></a> however&#44; some authors like Eric Topol<a class="elsevierStyleCrossRef" href="#bib0805"><span class="elsevierStyleSup">80</span></a> defend that black box procedures&#44; whose action mechanism is unknown&#44; are already accepted in medicine and&#44; therefore&#44; black box ML systems should be also accepted&#46; Moreover&#44; ML could provide us with a more deep understanding of sepsis&#44; opening up new ways to deal with it&#46;</p><p id="par0260" class="elsevierStylePara elsevierViewall">Another key aspect of ML in health care is the rigorous evaluation of the proposed models&#46; New ML-based systems for health care are proposed on a daily basis&#44; but many of them are retrospective studies&#46; Clinical prospective studies are more difficult to find and randomized controlled trials of AI systems are still an exception&#46; These trials pose challenges for patient or physician-level randomization&#44; since two different patient workflows should be used for the treatment and control groups&#44; which can be perceived as a risk and could be not be finally authorized by providers&#46;<a class="elsevierStyleCrossRef" href="#bib0795"><span class="elsevierStyleSup">78</span></a> In the case of sepsis&#44; the work of Shimabukuro et al&#46;<a class="elsevierStyleCrossRef" href="#bib0780"><span class="elsevierStyleSup">75</span></a> is virtually the unique case where a clinical trial has been carried out&#46;</p><p id="par0265" class="elsevierStylePara elsevierViewall">Other aspect to take into account is the road to the market&#44; which is probably not going to be easy&#44; as for traditional drugs&#46;<a class="elsevierStyleCrossRef" href="#bib0810"><span class="elsevierStyleSup">81</span></a></p><p id="par0270" class="elsevierStylePara elsevierViewall">Finally&#44; ethics plays a key-role in ML for health care&#44; also in sepsis&#46; A well-calibrated prognosis predictor on such critical condition could even raise dilemmas&#44; such as&#58; is it acceptable not to initiate or withdraw support measures in patients with a probability of dying above a certain threshold&#63; Could the results be artifacted by our prejudices when it comes to treating real patients&#63;</p><p id="par0275" class="elsevierStylePara elsevierViewall">ML is a very promising tool to improve sepsis detection and management&#44; but there is probably a long way in front of us and to go along it&#44; we will need to work as a team with partners unknown until now&#44; like AI experts&#46;</p></span><span id="sec0055" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0075">Author contribution</span><p id="par0280" class="elsevierStylePara elsevierViewall"><span class="elsevierStyleItalic">Initial concept and design</span>&#58; D&#46; Glez-Pe&#241;a&#44; P&#46; Vidal-Cort&#233;s&#46;</p><p id="par0285" class="elsevierStylePara elsevierViewall"><span class="elsevierStyleItalic">Clinical support for the research questions</span>&#58; P&#46; Vidal-Cort&#233;s&#44; L&#46; del R&#237;o Carbajo&#46;</p><p id="par0290" class="elsevierStylePara elsevierViewall"><span class="elsevierStyleItalic">Literature review and discussion&#58;</span> N&#46; Ocampo-Quintero&#44; M&#46; Reboiro-Jato&#44; F&#46; Fdez-Riverola&#46;</p><p id="par0295" class="elsevierStylePara elsevierViewall"><span class="elsevierStyleItalic">Drafting of the manuscript&#58;</span> N&#46; Ocampo-Quintero&#44; L&#46; del R&#237;o Carbajo&#44; F&#46; Fdez-Riverola&#44;</p><p id="par0300" class="elsevierStylePara elsevierViewall"><span class="elsevierStyleItalic">Final approval&#58;</span> N&#46; Ocampo-Quintero&#44; P&#46; Vidal-Cort&#233;s&#44; L&#46; del R&#237;o Carbajo&#44; F&#46; Fdez-Riverola&#44; M&#46; Reboiro-Jato&#44; D&#46; Glez-Pe&#241;a</p></span><span id="sec0060" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0080">Funding</span><p id="par0305" class="elsevierStylePara elsevierViewall">This work was partially supported by the <span class="elsevierStyleGrantSponsor" id="gs1">Conseller&#237;a de Educaci&#243;n&#44; Universidades e Formaci&#243;n Profesional &#40;Xunta de Galicia&#41;</span> under the scope of the strategic funding of ED431C2018&#47;55-GRC Competitive Reference Group&#46; N&#46; Ocampo-Quintero is supported by a doctoral scholarship from <span class="elsevierStyleGrantSponsor" id="gs2">National Council of Science and Technology &#40;CONACYT&#41;</span> &#40;member identification number&#58; CVU 681045&#41; from Mexico&#46;</p></span><span id="sec0065" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0085">Conflict of interests</span><p id="par0310" class="elsevierStylePara elsevierViewall">The authors have no conflict of interest to disclose&#46;</p></span></span>"
    "textoCompletoSecciones" => array:1 [
      "secciones" => array:19 [
        0 => array:3 [
          "identificador" => "xres1671963"
          "titulo" => "Abstract"
          "secciones" => array:1 [
            0 => array:1 [
              "identificador" => "abst0005"
            ]
          ]
        ]
        1 => array:2 [
          "identificador" => "xpalclavsec1484011"
          "titulo" => "Keywords"
        ]
        2 => array:3 [
          "identificador" => "xres1671964"
          "titulo" => "Resumen"
          "secciones" => array:1 [
            0 => array:1 [
              "identificador" => "abst0010"
            ]
          ]
        ]
        3 => array:2 [
          "identificador" => "xpalclavsec1484012"
          "titulo" => "Palabras clave"
        ]
        4 => array:2 [
          "identificador" => "sec0005"
          "titulo" => "Introduction"
        ]
        5 => array:2 [
          "identificador" => "sec0010"
          "titulo" => "Objectives"
        ]
        6 => array:2 [
          "identificador" => "sec0015"
          "titulo" => "Methodology"
        ]
        7 => array:2 [
          "identificador" => "sec0020"
          "titulo" => "Machine learning overview"
        ]
        8 => array:2 [
          "identificador" => "sec0025"
          "titulo" => "What are the key variables to build a clinical decision support system to assist in sepsis management&#63;"
        ]
        9 => array:2 [
          "identificador" => "sec0030"
          "titulo" => "Can sepsis be automatically detected earlier through the use of ML techniques&#63;"
        ]
        10 => array:2 [
          "identificador" => "sec0035"
          "titulo" => "Is sepsis prediction an achievable goal&#63;"
        ]
        11 => array:2 [
          "identificador" => "sec0040"
          "titulo" => "Can ML techniques be used to accurately predict sepsis-related mortality&#63;"
        ]
        12 => array:2 [
          "identificador" => "sec0045"
          "titulo" => "Is it possible to effectively increase adherence to treatment guidelines and reduce sepsis-related mortality and&#47;or associated costs using ML techniques&#63;"
        ]
        13 => array:2 [
          "identificador" => "sec0050"
          "titulo" => "Discussion and conclusions"
        ]
        14 => array:2 [
          "identificador" => "sec0055"
          "titulo" => "Author contribution"
        ]
        15 => array:2 [
          "identificador" => "sec0060"
          "titulo" => "Funding"
        ]
        16 => array:2 [
          "identificador" => "sec0065"
          "titulo" => "Conflict of interests"
        ]
        17 => array:2 [
          "identificador" => "xack588981"
          "titulo" => "Acknowledgements"
        ]
        18 => array:1 [
          "titulo" => "References"
        ]
      ]
    ]
    "pdfFichero" => "main.pdf"
    "tienePdf" => true
    "fechaRecibido" => "2019-11-29"
    "fechaAceptado" => "2020-04-05"
    "PalabrasClave" => array:2 [
      "en" => array:1 [
        0 => array:4 [
          "clase" => "keyword"
          "titulo" => "Keywords"
          "identificador" => "xpalclavsec1484011"
          "palabras" => array:4 [
            0 => "Sepsis"
            1 => "Clinical decision support systems"
            2 => "Machine learning"
            3 => "Artificial intelligence"
          ]
        ]
      ]
      "es" => array:1 [
        0 => array:4 [
          "clase" => "keyword"
          "titulo" => "Palabras clave"
          "identificador" => "xpalclavsec1484012"
          "palabras" => array:4 [
            0 => "Sepsis"
            1 => "Sistemas de apoyo a la decisi&#243;n cl&#237;nica"
            2 => "Aprendizaje autom&#225;tico"
            3 => "Inteligencia artificial"
          ]
        ]
      ]
    ]
    "tieneResumen" => true
    "resumen" => array:2 [
      "en" => array:2 [
        "titulo" => "Abstract"
        "resumen" => "<span id="abst0005" class="elsevierStyleSection elsevierViewall"><p id="spar0005" class="elsevierStyleSimplePara elsevierViewall">Sepsis is a major public health problem and a leading cause of death in the world&#44; where delay in the beginning of treatment&#44; along with clinical guidelines non-adherence have been proved to be associated with higher mortality&#46; Machine Learning is increasingly being adopted in developing innovative Clinical Decision Support Systems in many areas of medicine&#44; showing a great potential for automatic prediction of diverse patient conditions&#44; as well as assistance in clinical decision making&#46; In this context&#44; this work conducts a narrative review to provide an overview of how specific Machine Learning techniques can be used to improve sepsis management&#44; discussing the main tasks addressed&#44; the most popular methods and techniques&#44; as well as the obtained results&#44; in terms of both intelligent system accuracy and clinical outcomes improvement&#46;</p></span>"
      ]
      "es" => array:2 [
        "titulo" => "Resumen"
        "resumen" => "<span id="abst0010" class="elsevierStyleSection elsevierViewall"><p id="spar0010" class="elsevierStyleSimplePara elsevierViewall">La sepsis representa un problema de salud p&#250;blica de primer orden y es una de las principales causas de muerte a nivel mundial&#46; El retraso en el inicio del tratamiento&#44; junto con la no adherencia a las gu&#237;as de pr&#225;ctica cl&#237;nica se asocian a una mayor mortalidad&#46; El aprendizaje autom&#225;tico o <span class="elsevierStyleItalic">machine learning</span> est&#225;n siendo empleados en el desarrollo de sistemas de apoyo a la decisi&#243;n cl&#237;nica&#44; innovadores en muchas &#225;reas de la medicina&#44; mostrando un gran potencial para la predicci&#243;n de diversas condiciones del paciente&#44; as&#237; como en la asistencia durante el proceso de toma de decisiones m&#233;dicas&#46; En este sentido&#44; este trabajo lleva a cabo una revisi&#243;n narrativa para proporcionar una visi&#243;n general de c&#243;mo las t&#233;cnicas de <span class="elsevierStyleItalic">machine learning</span> pueden ser empleadas para mejorar el manejo de la sepsis&#44; discutiendo las principales tareas que tratan de resolver&#44; los m&#233;todos y las t&#233;cnicas m&#225;s empleados&#44; as&#237; como los resultados obtenidos&#44; tanto en t&#233;rminos de precisi&#243;n de los sistemas inteligentes&#44; como en la mejora de los resultados cl&#237;nicos&#46;</p></span>"
      ]
    ]
    "apendice" => array:1 [
      0 => array:1 [
        "seccion" => array:1 [
          0 => array:4 [
            "apendice" => "<p id="par0325" class="elsevierStylePara elsevierViewall">The following are the supplementary data to this article&#58;<elsevierMultimedia ident="upi0005"></elsevierMultimedia></p>"
            "etiqueta" => "Appendix A"
            "titulo" => "Supplementary data"
            "identificador" => "sec0075"
          ]
        ]
      ]
    ]
    "multimedia" => array:6 [
      0 => array:7 [
        "identificador" => "fig0005"
        "etiqueta" => "Figure 1"
        "tipo" => "MULTIMEDIAFIGURA"
        "mostrarFloat" => true
        "mostrarDisplay" => false
        "figura" => array:1 [
          0 => array:4 [
            "imagen" => "gr1.jpeg"
            "Alto" => 2841
            "Ancho" => 2917
            "Tamanyo" => 399170
          ]
        ]
        "descripcion" => array:1 [
          "en" => "<p id="spar0015" class="elsevierStyleSimplePara elsevierViewall">Graphic representation of simple classifiers from different families&#58; &#40;a&#41; an ANN classifier with one hidden layer&#44; &#40;b&#41; a bi-dimensional SVM&#44; &#40;c&#41; a LR with one input variable&#44; and &#40;d&#41; a simple DT&#46;</p>"
        ]
      ]
      1 => array:8 [
        "identificador" => "tbl0005"
        "etiqueta" => "Table 1"
        "tipo" => "MULTIMEDIATABLA"
        "mostrarFloat" => true
        "mostrarDisplay" => false
        "detalles" => array:1 [
          0 => array:3 [
            "identificador" => "at1"
            "detalle" => "Table "
            "rol" => "short"
          ]
        ]
        "tabla" => array:1 [
          "tablatextoimagen" => array:1 [
            0 => array:2 [
              "tabla" => array:1 [
                0 => """
                  <table border="0" frame="\n
                  \t\t\t\t\tvoid\n
                  \t\t\t\t" class=""><thead title="thead"><tr title="table-row"><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Task&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">ML technique&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">References&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th></tr></thead><tbody title="tbody"><tr title="table-row"><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " rowspan="5" align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis detection</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Artificial Neural Networks&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;36&#41; &#40;37&#41; &#40;38&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Bayesian&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;39&#41; &#40;40&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Decision Trees&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;39&#41; &#40;41&#41; &#40;42&#41; &#40;43&#41; &#40;44&#41; &#40;45&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Logistic Regressions&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;46&#41; &#40;47&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Support Vector Machines&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;39&#41; &#40;48&#41; &#40;49&#41; &#40;50&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " colspan="3" align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t"><span class="elsevierStyleVsp" style="height:0.5px"></span></td></tr><tr title="table-row"><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " rowspan="3" align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis prediction</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Artificial Neural Networks&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;37&#41; &#40;38&#41; &#40;51&#41; &#40;52&#41; &#40;53&#41; &#40;54&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Decision Trees&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;41&#41; &#40;55&#41; &#40;56&#41; &#40;45&#41; &#40;57&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Other Techniques&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;55&#41; &#40;58&#41; &#40;51&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " colspan="3" align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t"><span class="elsevierStyleVsp" style="height:0.5px"></span></td></tr><tr title="table-row"><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " rowspan="6" align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Mortality prediction</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Artificial Neural Networks&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;59&#41; &#40;60&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Bayesian&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;61&#41; &#40;62&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Decision Trees&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;63&#41; &#40;64&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Logistic Regressions&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;65&#41; &#40;59&#41; &#40;63&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Support Vector Machines&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;59&#41; &#40;61&#41; &#40;66&#41; &#40;64&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Other techniques&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;67&#41; &#40;64&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr></tbody></table>
                  """
              ]
              "imagenFichero" => array:1 [
                0 => "xTab2844747.png"
              ]
            ]
          ]
        ]
        "descripcion" => array:1 [
          "en" => "<p id="spar0020" class="elsevierStyleSimplePara elsevierViewall">Main tasks comprising sepsis management with reference to ML approaches used by the reviewed studies&#46;</p>"
        ]
      ]
      2 => array:8 [
        "identificador" => "tbl0010"
        "etiqueta" => "Table 2"
        "tipo" => "MULTIMEDIATABLA"
        "mostrarFloat" => true
        "mostrarDisplay" => false
        "detalles" => array:1 [
          0 => array:3 [
            "identificador" => "at2"
            "detalle" => "Table "
            "rol" => "short"
          ]
        ]
        "tabla" => array:3 [
          "leyenda" => "<p id="spar0030" class="elsevierStyleSimplePara elsevierViewall">Table&#39;s acronyms and abbreviations&#58;</p><p id="spar0035" class="elsevierStyleSimplePara elsevierViewall">AUC&#58; area under curve&#59; CI&#58; confidence interval&#59; BoW&#58; bag of words&#59; CC&#58; classifier chains&#59; AUROC&#58; area under the receiver operating characteristic curve&#59; DBN&#58; dynamic Bayesian network&#59; DFN&#58; deep feedforward networks&#59; GTB&#58; gradient tree boosting&#59; DT&#58; decision tree&#59; LR&#58; logistic regression&#59; LSTM&#58; long short term memory&#59; KELM&#58; kernel extreme learning machine&#59; ML&#58; machine learning&#59; MGP&#58; multiple-output gaussian process&#59; NB&#58; na&#239;ve bayes&#59; RNN&#58; recurrent neural network&#59; SVM&#58; support vector machines&#46;</p><p id="spar0040" class="elsevierStyleSimplePara elsevierViewall">DBP&#58; diastolic blood pressure&#59; Ear-PPG&#58; ear photoplethysmography&#59; ECG&#58; electrocardiogram&#44; ED&#58; emergency department&#59; ICU&#58; intensive care unit&#59; SIRS&#58; systemic inflammatory response syndrome&#59; Fin-PPG&#58; finger photoplethysmography&#59; GCS&#58; glasgow coma scale&#59; HR&#58; heart rate&#59; LOS&#58; length of stay&#59; MBP&#58; mean blood pressure&#59; PaCO<span class="elsevierStyleInf">2</span>&#58; partial pressure of carbon dioxide&#59; PP&#58; pulse pressure&#59; RR&#58; respiratory rate&#59; SaO<span class="elsevierStyleInf">2</span>&#58; oxygen arterial saturation&#59; SBP&#58; systolic blood pressure&#59; T&#176;&#58; temperature&#59; WBC&#58; white blood cells&#46;</p><p id="spar0045" class="elsevierStyleSimplePara elsevierViewall">NS&#58; no specified&#46;</p><p id="spar0050" class="elsevierStyleSimplePara elsevierViewall">Index time&#46;</p>"
          "tablatextoimagen" => array:1 [
            0 => array:2 [
              "tabla" => array:1 [
                0 => """
                  <table border="0" frame="\n
                  \t\t\t\t\tvoid\n
                  \t\t\t\t" class=""><thead title="thead"><tr title="table-row"><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Ref&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">ML type&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Variables&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">AUC &#40;CI 95&#37;&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Sensitivity &#40;Sen&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Specificity &#40;Sp&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th></tr></thead><tbody title="tbody"><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Mao et al&#46;&#44; 2018<a class="elsevierStyleCrossRef" href="#bib0615"><span class="elsevierStyleSup">42</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">GTB&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">6 &#40;SaO<span class="elsevierStyleInf">2</span>&#44; HR&#44; SBP&#44; DBP&#44; T&#176;&#44; and RR&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ED&#44; hospital wards and ICU patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis&#58; 0&#46;92 &#40;0&#46;90&#8211;0&#46;93&#41;Severe sepsis&#58; 0&#46;87 &#40;0&#46;86&#8211;0&#46;88&#41;Septic Shock&#58; 0&#46;9992 &#40;0&#46;9991&#8211;0&#46;9994&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;Sen fixed near 0&#46;80&#41;&#58;Sepsis&#58; 0&#46;95 &#40;0&#46;93&#8211;0&#46;97&#41;Severe sepsis&#58; 0&#46;85 &#40;0&#46;84&#8211;0&#46;86&#41;Septic shock&#58; 0&#46;9990 &#40;0&#46;9987&#8211;0&#46;9993&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;Sp fixed near 0&#46;80&#41;&#58;Sepsis&#58; 0&#46;98 &#40;0&#46;96&#8211;1&#46;00&#41;Severe sepsis&#58; 0&#46;996 &#40;0&#46;989&#8211;1&#46;00&#41;Septic shock&#58; 1&#46;00 &#40;1&#46;00&#8211;1&#46;00&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Gon&#231;alves et al&#46;&#44; 2013<a class="elsevierStyleCrossRef" href="#bib0600"><span class="elsevierStyleSup">39</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">DTNBSVM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">9 &#40;bilirubin&#44; creatinine&#44; glucose&#44; leukocytes&#44; platelets&#44; HR&#44; MBP&#44; SBP&#44; and T&#176;&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">DT&#58; 1&#46;00NB&#58; 0&#46;9982SVM&#58; 1&#46;00&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">DT&#58; 1&#46;00NB&#58; 1&#46;00SVM&#58; 1&#46;00&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">DT&#58; 1&#46;00NB&#58; 0&#46;9990SVM&#58; 1&#46;00&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Futoma et al&#46;&#44; 2017<a class="elsevierStyleCrossRef" href="#bib0595"><span class="elsevierStyleSup">38</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">MGP-RNN &#40;LSTM&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">77 &#40;34 physiological variables &#40;6 vital signs&#44; 28 laboratory values&#41;&#44;35 covariates &#40;29 comorbidities and other 6&#41;&#44; and 8 medication classes&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Hospital&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#62;0&#46;90&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Horng et al&#46;&#44; 2017<a class="elsevierStyleCrossRef" href="#bib0645"><span class="elsevierStyleSup">48</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">SVM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">12 &#40;age&#44; gender&#44; acuity&#44; SBP&#44; DBP&#44; HR&#44; Pain Scale&#44; RR&#44; SaO<span class="elsevierStyleInf">2</span>&#44; T&#176;&#44; free text chief complaint&#44; and free text nursing assessments&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ED patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Vitals&#58; 0&#46;67CC&#58; 0&#46;83BoW&#58; 0&#46;86Topics&#58; 0&#46;85&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Vitals&#58; 0&#46;56CC&#58; 0&#46;75BoW&#58; 0&#46;78Topics&#58; 0&#46;80&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Vitals&#58; 0&#46;68CC&#58; 0&#46;75BoW&#58; 0&#46;79Topics&#58; 0&#46;75&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Tang et al&#46;&#44; 2010<a class="elsevierStyleCrossRef" href="#bib0650"><span class="elsevierStyleSup">49</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">SVM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">3 &#40;Non-invasive cardiovascular variables&#58; ECG&#44; Fin-PPG&#44; and Ear-PPG&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ED patients with SIRS&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Severe sepsis&#58; 0&#46;78&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;9444&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;6250&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Faisal et al&#46;&#44; 2018<a class="elsevierStyleCrossRef" href="#bib0640"><span class="elsevierStyleSup">47</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LR&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">19 &#40;First electronically recorded vital signs and blood test results&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ED patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis&#58; 0&#46;7908Severe sepsis&#58; 0&#46;9036&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis&#58; 0&#46;5434Severe sepsis&#58;0&#46;5306&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Wang X&#46; et al&#46;&#44; 2018<a class="elsevierStyleCrossRef" href="#bib0585"><span class="elsevierStyleSup">36</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">KELM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">5 &#40;<span class="elsevierStyleSmallCaps">d</span>-xylose&#44; acetatic acid&#44; linoleic acid&#44; <span class="elsevierStyleSmallCaps">d</span>-glucopyranosiduronic acid&#44; and cholesterol&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ED and ICU patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;8957&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;6577&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Nachimuthu et al&#46;&#44; 2012<a class="elsevierStyleCrossRef" href="#bib0605"><span class="elsevierStyleSup">40</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">DBN&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">10 &#40;WBC&#44; &#37; of immature neutrophiles&#44; HR&#44; MBP&#44; DBP&#44; SBP&#44; T&#176;&#44; RR&#44; PaCO<span class="elsevierStyleInf">2</span>&#44; and age&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ED patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">First 3<span class="elsevierStyleHsp" style=""></span>h&#58; 0&#46;91102First 6<span class="elsevierStyleHsp" style=""></span>h&#58; 0&#46;91499First 12<span class="elsevierStyleHsp" style=""></span>h&#58; 0&#46;93362First 24<span class="elsevierStyleHsp" style=""></span>h&#58; 0&#46;94353&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">First 3<span class="elsevierStyleHsp" style=""></span>h&#58; 0&#46;68902First 6<span class="elsevierStyleHsp" style=""></span>h&#58; 0&#46;70732First 12<span class="elsevierStyleHsp" style=""></span>h&#58; 0&#46;81707First 24<span class="elsevierStyleHsp" style=""></span>h&#58; 0&#46;85976&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">First 3<span class="elsevierStyleHsp" style=""></span>h&#58; 0&#46;94881First 6<span class="elsevierStyleHsp" style=""></span>h&#58; 0&#46;94994First 12<span class="elsevierStyleHsp" style=""></span>h&#58; 0&#46;94881First 24<span class="elsevierStyleHsp" style=""></span>h&#58; 0&#46;94539&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Desautels et al&#46;&#44; 2016<a class="elsevierStyleCrossRef" href="#bib0610"><span class="elsevierStyleSup">41</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">GTB&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">8 &#40;age&#44; HR&#44; DBP&#44; SBP&#44; T&#176;&#44; RR&#44; SaO<span class="elsevierStyleInf">2</span>&#44; and GCS&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;88<span class="elsevierStyleHsp" style=""></span>&#177;<span class="elsevierStyleHsp" style=""></span>0&#46;006&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;80&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;80&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Kam&#44; H&#46;J&#46; and Kim&#44; H&#46;Y&#46;&#44; 2017<a class="elsevierStyleCrossRef" href="#bib0590"><span class="elsevierStyleSup">37</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">DFNLSTM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">9 &#40;pH&#44; SaO<span class="elsevierStyleInf">2</span>&#44; WBC&#44; HR&#44; SBP&#44; PP&#44; T&#176;&#44; RR&#44; and age&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">SIRS criteria &#40;hospital and ICU&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LSTM&#58; 0&#46;99&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LSTM&#58; 0&#46;97&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LSTM&#58; 1&#46;00&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Back et al&#46;&#44; 2016<a class="elsevierStyleCrossRef" href="#bib0635"><span class="elsevierStyleSup">46</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LR&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">7 &#40;HR&#44; DBP&#44; T&#176;&#44; RR&#44; age&#44; admission via ED&#44; LOS&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Hospital&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Test&#47;Validation0&#46;96&#47;0&#46;95&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Test&#47;Validation0&#46;96&#47;0&#46;77&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Test&#47;Validation0&#46;83&#47;0&#46;96&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Arvind et al&#46;&#44; 2018<a class="elsevierStyleCrossRef" href="#bib0655"><span class="elsevierStyleSup">50</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">SVM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Clinical notes&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU post-surgical patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;82&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;79&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;85&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Delahanty et al&#46; 2019<a class="elsevierStyleCrossRef" href="#bib0620"><span class="elsevierStyleSup">43</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">GTB&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">13 &#40;8 laboratory results&#44; 3 vital signs&#44; 2 &#8220;engineered&#8221;&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ED patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Time after an index time<a class="elsevierStyleCrossRef" href="#tblfn0005"><span class="elsevierStyleSup">a</span></a>&#58;0&#46;93 at 1<span class="elsevierStyleHsp" style=""></span>h0&#46;95 at 3<span class="elsevierStyleHsp" style=""></span>h0&#46;96 at 6<span class="elsevierStyleHsp" style=""></span>h0&#46;97 at 12<span class="elsevierStyleHsp" style=""></span>h0&#46;97 at 24<span class="elsevierStyleHsp" style=""></span>h&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Time after an index time<a class="elsevierStyleCrossRef" href="#tblfn0005"><span class="elsevierStyleSup">a</span></a>&#58;0&#46;68 at 1<span class="elsevierStyleHsp" style=""></span>h0&#46;72 at 3<span class="elsevierStyleHsp" style=""></span>h0&#46;75 at 6<span class="elsevierStyleHsp" style=""></span>h0&#46;79 at 12<span class="elsevierStyleHsp" style=""></span>h0&#46;85 at 24<span class="elsevierStyleHsp" style=""></span>h&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Time after an index time<a class="elsevierStyleCrossRef" href="#tblfn0005"><span class="elsevierStyleSup">a</span></a>&#58;0&#46;96 at 1<span class="elsevierStyleHsp" style=""></span>h0&#46;97 at 3<span class="elsevierStyleHsp" style=""></span>h0&#46;97 at 6<span class="elsevierStyleHsp" style=""></span>h0&#46;96 at 12<span class="elsevierStyleHsp" style=""></span>h0&#46;96 at 24<span class="elsevierStyleHsp" style=""></span>h&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Calvert et al&#46;&#44; 2019<a class="elsevierStyleCrossRef" href="#bib0625"><span class="elsevierStyleSup">44</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">GTB&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">6 &#40;DBP&#44; SBP&#44; HR&#44; T&#176;&#44; RR&#44; SpO<span class="elsevierStyleInf">2</span>&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">High-risk patients &#40;age<span class="elsevierStyleHsp" style=""></span>&#8805;<span class="elsevierStyleHsp" style=""></span>45 years and length-of-stay<span class="elsevierStyleHsp" style=""></span>&#8805;<span class="elsevierStyleHsp" style=""></span>4 days&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;917&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;799&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;860&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Barton et al&#46;&#44; 2019<a class="elsevierStyleCrossRef" href="#bib0630"><span class="elsevierStyleSup">45</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">GTB&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">6 &#40;SaO<span class="elsevierStyleInf">2</span>&#44; HR&#44; SBP&#44; DBP&#44; T&#176;&#44; and RR&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Hospital&#44; ICU&#44; ED patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;88&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;80&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;78&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr></tbody></table>
                  """
              ]
              "imagenFichero" => array:1 [
                0 => "xTab2844749.png"
              ]
            ]
          ]
          "notaPie" => array:1 [
            0 => array:3 [
              "identificador" => "tblfn0005"
              "etiqueta" => "a"
              "nota" => "<p class="elsevierStyleNotepara" id="npar0005">The time which the first vital sign or laboratory result was documented in the EHR&#46;</p>"
            ]
          ]
        ]
        "descripcion" => array:1 [
          "en" => "<p id="spar0025" class="elsevierStyleSimplePara elsevierViewall">Main characteristics of the ML classifiers used for sepsis detection&#46;</p>"
        ]
      ]
      3 => array:8 [
        "identificador" => "tbl0015"
        "etiqueta" => "Table 3"
        "tipo" => "MULTIMEDIATABLA"
        "mostrarFloat" => true
        "mostrarDisplay" => false
        "detalles" => array:1 [
          0 => array:3 [
            "identificador" => "at3"
            "detalle" => "Table "
            "rol" => "short"
          ]
        ]
        "tabla" => array:2 [
          "leyenda" => "<p id="spar0060" class="elsevierStyleSimplePara elsevierViewall">Table&#39;s acronyms and abbreviations&#58;</p><p id="spar0065" class="elsevierStyleSimplePara elsevierViewall">AUC&#58; area under curve&#59; CI&#58; confidence interval&#59; DFN&#58; deep feedforward networks&#59; GTB&#58; gradient tree boosting&#59; LiR&#58; linear regression&#59; LR&#58; logistic regression&#59; LMT&#58; logistic model tree&#59; LSTM&#58; long short term memory&#59; GLM&#58; generalized linear model&#59; XGBoost&#58; extreme gradient boosting&#59; ML&#58; machine learning&#59; MGP&#58; multiple-output gaussian process&#59; MLP&#58; multilayer perceptron&#59; RF&#58; random forest&#59; RNN&#58; recurrent neural network&#59; SVM&#58; support vector machines&#46;</p><p id="spar0070" class="elsevierStyleSimplePara elsevierViewall">BUN&#58; blood urea nitrogen&#59; Ca&#58; Calcium&#59; DBP&#58; diastolic blood pressure&#59; Ear-PPG&#58; ear photoplethysmography&#59; ED&#58; emergency department&#59; ICU&#58; intensive care unit&#59; SIRS&#58; systemic inflammatory response syndrome&#59; GCS&#58; glasgow coma scale&#59; Hb&#58; haemoglobin&#59; HR&#58; heart rate&#59; INR&#58; international normalized ratio&#59; LOS&#58; length of stay&#59; Mg&#58; magnesium&#59; MBP&#58; mean blood pressure&#59; P&#58; phosphorus&#59; PaCO<span class="elsevierStyleInf">2</span>&#58; partial pressure of carbon dioxide&#59; PP&#58; pulse pressure&#59; RR&#58; respiratory rate&#59; SaO<span class="elsevierStyleInf">2</span>&#58; oxygen arterial saturation&#59; SBP&#58; systolic blood pressure&#59; T&#176;&#58; temperature&#59; WBC&#58; white blood cells&#46;</p><p id="spar0075" class="elsevierStyleSimplePara elsevierViewall">NS&#58; no specified&#46;</p>"
          "tablatextoimagen" => array:1 [
            0 => array:2 [
              "tabla" => array:1 [
                0 => """
                  <table border="0" frame="\n
                  \t\t\t\t\tvoid\n
                  \t\t\t\t" class=""><thead title="thead"><tr title="table-row"><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Ref&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">ML type&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Variables&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Objectives&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">AUC &#40;CI 95&#37;&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Sensitivity &#40;Sen&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Specificity &#40;Sp&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th></tr></thead><tbody title="tbody"><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Desautels et al&#46;&#44; 2016<a class="elsevierStyleCrossRef" href="#bib0610"><span class="elsevierStyleSup">41</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">GTB&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">8 &#40;age&#44; HR&#44; DBP&#44; SBP&#44; T&#176;&#44; RR&#44; SaO<span class="elsevierStyleInf">2</span>&#44; and GCS&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">4<span class="elsevierStyleHsp" style=""></span>h before sepsis&#58; 0&#46;74<span class="elsevierStyleHsp" style=""></span>&#177;<span class="elsevierStyleHsp" style=""></span>0&#46;010&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;80&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;54&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Futoma et al&#46;&#44; 2017<a class="elsevierStyleCrossRef" href="#bib0595"><span class="elsevierStyleSup">38</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">MGP-RNN &#40;LSTM&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">77 &#40;34 physiological variables&#44; 35 covariates and 8 medication classes&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Hospital&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">2<span class="elsevierStyleHsp" style=""></span>h before&#58; &#8764;0&#46;874<span class="elsevierStyleHsp" style=""></span>h before&#58;&#8764;0&#46;846<span class="elsevierStyleHsp" style=""></span>h before&#58; &#8764;0&#46;8212<span class="elsevierStyleHsp" style=""></span>h before&#58; &#8764;0&#46;77&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Kam&#44; H&#46;J&#46; and Kim&#44; H&#46;Y&#46;&#44; 2017<a class="elsevierStyleCrossRef" href="#bib0590"><span class="elsevierStyleSup">37</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">DFNLSTM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">9 &#40;pH&#44; SaO<span class="elsevierStyleInf">2</span>&#44; WBC&#44; HR&#44; SBP&#44; PP&#44; T&#176;&#44; RR&#44; and age&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">SIRS criteria &#40;hospital and ICU&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LSTM 1<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;96LSTM 2<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;94LSTM 3<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;929DFN100 3<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;915&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LSTM 1<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;92LSTM 2<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;89LSTM 3<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;914DFN100 3<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;886&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LSTM 1<span class="elsevierStyleHsp" style=""></span>h before&#58; 1&#46;00LSTM 2<span class="elsevierStyleHsp" style=""></span>h before&#58; 1&#46;00LSTM 3<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;944DFN100 3<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;944&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Wyk et al&#46;&#44; 2018<a class="elsevierStyleCrossRef" href="#bib0690"><span class="elsevierStyleSup">57</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">RF&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">8 &#40;HR&#44; RR&#44; SBP&#44; DBP&#44; T&#176;&#44; SpO2&#44; WBC&#44; LOS&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">1<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;7&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">1<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;8&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">1<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;6&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Nemati et al&#46;&#44; 2018<a class="elsevierStyleCrossRef" href="#bib0695"><span class="elsevierStyleSup">58</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Weilbull-Cox Hazards Model&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">65 &#40;30 laboratory values&#44; 6 high-resolution dynamical features&#44; 10 clinical features&#44; 19 demographics&#47;context features&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">4<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;846<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;828<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;8212<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;79&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Fixed at 0&#46;85&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">4<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;646<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;628<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;6212<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;57&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Wang&#44; R&#46;Z&#46; et al&#46;&#44; 2018<a class="elsevierStyleCrossRef" href="#bib0680"><span class="elsevierStyleSup">55</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LRSVMLMT&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">12 &#40;P&#44; Ca&#44; Mg&#44; BUN&#44; Hb&#44; Platelets&#44; WBC&#44; INR&#44; Alkaline Phosphatase&#44; HR&#44; SBP&#44; and SaO<span class="elsevierStyleInf">2</span>&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">6<span class="elsevierStyleHsp" style=""></span>h before&#58;LR&#58; 0&#46;685SVM&#58; 0&#46;674LMT&#58; 0&#46;750&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">6<span class="elsevierStyleHsp" style=""></span>h before&#58;LR&#58; 0&#46;752SVM&#58; 0&#46;566LMT&#58; 0&#46;671&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">6<span class="elsevierStyleHsp" style=""></span>h before&#58;LR&#58; 0&#46;618SVM&#58; 0&#46;783LMT&#58; 0&#46;830&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Giannini et al&#46;2019<a class="elsevierStyleCrossRef" href="#bib0685"><span class="elsevierStyleSup">56</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">RF&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">587 &#40;demographics&#44; vital signs and laboratory results&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Hospital wards&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;88&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;26&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;98&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Schamoni&#46; et al&#46;&#44; 2019<a class="elsevierStyleCrossRef" href="#bib0660"><span class="elsevierStyleSup">51</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LiRMLP&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">57 &#40;42 vital signs and laboratory results&#44; 3 demographic&#44; 10 pre-existing conditions&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LiR&#58;0&#46;808 &#40;0&#46;786&#8211;0&#46;830&#41; 12&#8211;8<span class="elsevierStyleHsp" style=""></span>h before sepsis onset0&#46;770 &#40;0&#46;739&#8211;0&#46;801&#41; 24&#8211;12<span class="elsevierStyleHsp" style=""></span>hMLP&#58;0&#46;817 &#40;0&#46;789&#8211;0&#46;844&#41; 12&#8211;8<span class="elsevierStyleHsp" style=""></span>h before0&#46;776 &#40;0&#46;739&#8211;0&#46;811&#41; 24&#8211;12<span class="elsevierStyleHsp" style=""></span>h before&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Barton et al&#46;&#44; 2019<a class="elsevierStyleCrossRef" href="#bib0630"><span class="elsevierStyleSup">45</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">GTB&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">6 &#40;SaO<span class="elsevierStyleInf">2</span>&#44; HR&#44; SBP&#44; DBP&#44; T&#176;&#44; and RR&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Hospital&#44; ICU&#44; ED patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;84 24<span class="elsevierStyleHsp" style=""></span>h before sepsis onset0&#46;83 48<span class="elsevierStyleHsp" style=""></span>h before sepsis onset&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;80 24<span class="elsevierStyleHsp" style=""></span>h before sepsis onset0&#46;84 48<span class="elsevierStyleHsp" style=""></span>h before sepsis onset&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;72 24<span class="elsevierStyleHsp" style=""></span>h before sepsis onset0&#46;66 48<span class="elsevierStyleHsp" style=""></span>h before sepsis onset&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Scherpf et al&#46;&#44; 2019<a class="elsevierStyleCrossRef" href="#bib0665"><span class="elsevierStyleSup">52</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">RNN&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">10 &#40;age&#44; DBP&#44; SBP&#44; pH&#44; SaO2&#44; T&#176;&#44; HR&#44; RR&#44; PaCO<span class="elsevierStyleInf">2</span>&#44; WBC&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">3<span class="elsevierStyleHsp" style=""></span>h before&#58; 0&#46;81 &#40;0&#46;78&#8211;0&#46;84&#41;&#46;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;Sp fixed&#58; 0&#46;90&#41;3<span class="elsevierStyleHsp" style=""></span>h before&#58; 47&#46;06<span class="elsevierStyleHsp" style=""></span>h before&#58; 44&#46;912<span class="elsevierStyleHsp" style=""></span>h before&#58; 46&#46;3&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#40;Sen fixed at 0&#46;90&#41;47&#46;0 &#40;95&#37; CI&#58; 43&#46;1&#37;&#8211;50&#46;8&#37;&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Kaji et al&#46;&#44; 2019<a class="elsevierStyleCrossRef" href="#bib0670"><span class="elsevierStyleSup">53</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LSTMRNN&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">119 &#40;demographic data&#44; vitals&#44; labs&#44; and treatment&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis Prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Same-day sepsis&#58; 0&#46;952Next-day sepsis&#58; 0&#46;876&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Same-day sepsis&#58; 0&#46;73Next-day sepsis&#58; 0&#46;57&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Fagerstr&#246;m et al&#46;&#44; 2019<a class="elsevierStyleCrossRef" href="#bib0675"><span class="elsevierStyleSup">54</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LSTM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">24 &#40;demographic data&#44; vital signs&#44; laboratory results&#44; treatment&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;83 &#40;48<span class="elsevierStyleHsp" style=""></span>h before&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Mao et al&#46;&#44; 2018<a class="elsevierStyleCrossRef" href="#bib0615"><span class="elsevierStyleSup">42</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">GTB&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">6 &#40;SaO<span class="elsevierStyleInf">2</span>&#44; HR&#44; SBP&#44; DBP&#44; T&#176;&#44; and RR&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis severity prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Septic patients in ED&#44; hospital wards and ICU&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">4<span class="elsevierStyleHsp" style=""></span>h beforeSevere sepsis&#58; 0&#46;85 &#40;0&#46;79&#8211;0&#46;91&#41;Septic shock&#58; 0&#46;96 &#40;0&#46;94&#8211;0&#46;98&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Lin et al&#46;&#44; 2018<a class="elsevierStyleCrossRef" href="#bib0750"><span class="elsevierStyleSup">69</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LSTM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">43 &#40;6 vital signs&#44; 11 laboratory values&#44; 4 treatment&#44; 18 culture results and 4 other&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Sepsis severity prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Hospital patients with suspected infection&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">12<span class="elsevierStyleHsp" style=""></span>h before septic shock&#58; 0&#46;9411&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">12<span class="elsevierStyleHsp" style=""></span>h before septic shock&#58; 0&#46;8408&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Liu et al&#46;&#44; 2019<a class="elsevierStyleCrossRef" href="#bib0755"><span class="elsevierStyleSup">70</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">GLMXGBoostRNN&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">32 &#40;13 vital signs&#44; 12 laboratory results&#44; 7 treatment&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Septic shock prediction&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU patients with suspected infection&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">GLM&#58; 0&#46;82XGBoost&#58; 0&#46;83RNN&#58; 0&#46;85&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">GLM&#58; 0&#46;85XGBoost&#58; 0&#46;77RNN&#58; 0&#46;79&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">GLM&#58; 0&#46;73XGBoost&#58; 0&#46;73RNN&#58; 0&#46;77&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " colspan="3" align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Median early warning time&#58;GLM&#58; 9&#46;5<span class="elsevierStyleHsp" style=""></span>hXGBoost&#58; 9&#46;0<span class="elsevierStyleHsp" style=""></span>hRNN&#58; 10&#46;3<span class="elsevierStyleHsp" style=""></span>h</td></tr></tbody></table>
                  """
              ]
              "imagenFichero" => array:1 [
                0 => "xTab2844748.png"
              ]
            ]
          ]
        ]
        "descripcion" => array:1 [
          "en" => "<p id="spar0055" class="elsevierStyleSimplePara elsevierViewall">Main characteristics of the ML classifiers used for sepsis prediction&#46;</p>"
        ]
      ]
      4 => array:8 [
        "identificador" => "tbl0020"
        "etiqueta" => "Table 4"
        "tipo" => "MULTIMEDIATABLA"
        "mostrarFloat" => true
        "mostrarDisplay" => false
        "detalles" => array:1 [
          0 => array:3 [
            "identificador" => "at4"
            "detalle" => "Table "
            "rol" => "short"
          ]
        ]
        "tabla" => array:2 [
          "leyenda" => "<p id="spar0085" class="elsevierStyleSimplePara elsevierViewall">Table&#39;s acronyms and abbreviations&#58;</p><p id="spar0090" class="elsevierStyleSimplePara elsevierViewall">AUC&#58; area under curve&#59; CI&#58; confidence interval&#59; AB&#58; adaptive boosting&#59; AE&#58; auto encoder&#59; CART&#58; classification and regression tree&#59; CNN&#58; convolutional neural network&#59; DBN&#58; dynamic Bayesian network&#59; GTB&#58; gradient tree boosting&#59; LASSO&#58; least absolute shrinkage and selection operator&#59; SGB&#58; stochastic gradient boosting&#59; LR&#58; logistic regression&#59; kNN&#58; k-nearest neighbours&#59; ML&#58; machine learning&#59; MLP&#58; multilayer perceptron&#59; NB&#58; na&#239;ve bayes&#59; PCA&#58; principal component analysis&#59; RF&#58; random forest&#59; SVM&#58; support vector machines&#46;</p><p id="spar0095" class="elsevierStyleSimplePara elsevierViewall">Ca&#58; calcium&#59; DBP&#58; diastolic blood pressure&#59; Ear-PPG&#58; ear photoplethysmography&#59; ED&#58; emergency department&#59; ICU&#58; intensive care unit&#59; SIRS&#58; systemic inflammatory response syndrome&#59; GCS&#58; glasgow coma scale&#59; HRV&#58; heart rate variability&#59; MBP&#58; mean blood pressure&#59; RR&#58; respiratory rate&#59; T&#176;&#58; temperature&#59; WBC&#58; white blood cells&#46;</p>"
          "tablatextoimagen" => array:1 [
            0 => array:2 [
              "tabla" => array:1 [
                0 => """
                  <table border="0" frame="\n
                  \t\t\t\t\tvoid\n
                  \t\t\t\t" class=""><thead title="thead"><tr title="table-row"><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Ref&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">ML type&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Variables&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">AUC &#40;CI 95&#37;&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Sensitivity &#40;Sen&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th><th class="td" title="\n
                  \t\t\t\t\ttable-head\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t" scope="col" style="border-bottom: 2px solid black">Specificity &#40;Sp&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t\t\t</th></tr></thead><tbody title="tbody"><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Ribas et al&#46;&#44; 2012<a class="elsevierStyleCrossRef" href="#bib0730"><span class="elsevierStyleSup">65</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LR&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">34 &#40;demographic&#44; comorbidities&#44; organ function&#44; treatment&#44; infection&#41;&#46; First 24<span class="elsevierStyleHsp" style=""></span>h of evolution&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Severe sepsis admitted to ICU&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;75&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;64&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;84&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Gultepe et al&#46;&#44; 2014<a class="elsevierStyleCrossRef" href="#bib0710"><span class="elsevierStyleSup">61</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">NBSVM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">5 &#40;Lactate&#44; MBP&#44; RR&#44; T&#176;&#44; and WBC&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Hospital patients with SIRS&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">NB&#58; 0&#46;660<span class="elsevierStyleHsp" style=""></span>&#177;<span class="elsevierStyleHsp" style=""></span>0&#46;050SVM&#58; 0&#46;726<span class="elsevierStyleHsp" style=""></span>&#177;<span class="elsevierStyleHsp" style=""></span>0&#46;045&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">NB&#58; 0&#46;879SVM&#58; 0&#46;949&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">NB&#58; 0&#46;385SVM&#58; 0&#46;308&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Tsoukalas et al&#46;&#44; 2015<a class="elsevierStyleCrossRef" href="#bib0735"><span class="elsevierStyleSup">66</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">SVM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">5 &#40;T&#176;&#44; RR&#44; WBC&#44; MBP&#44; and lactate&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Patients with SIRS&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;61&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Taylor et al&#46;&#44; 2016<a class="elsevierStyleCrossRef" href="#bib0720"><span class="elsevierStyleSup">63</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">RFCARTLR&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#62;500 &#40;demographic&#44; previous health status&#44; ED health status&#44; ED services render&#44; and operational details&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ED septic patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">28 days mortalityRF&#58; 0&#46;860 &#40;0&#46;819&#8211;0&#46;900&#41;CART&#58; 0&#46;693 &#40;0&#46;620&#8211;0&#46;766&#41;LR&#58; 0&#46;755 &#40;0&#46;689&#8211;0&#46;821&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Byrne et al&#46;&#44; 2016<a class="elsevierStyleCrossRef" href="#bib0700"><span class="elsevierStyleSup">59</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LRMLPSVM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">939 peptides from LC-MS&#47;MS at<span class="elsevierStyleHsp" style=""></span>&#60;<span class="elsevierStyleHsp" style=""></span>16<span class="elsevierStyleHsp" style=""></span>h and 48<span class="elsevierStyleHsp" style=""></span>h after septic shock diagnosis&#46;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU patients with septic shock&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#60;16<span class="elsevierStyleHsp" style=""></span>h after shock diagnosis&#58;LR&#58; 0&#46;7415MLP&#58; 0&#46;7222SVM&#58; 0&#46;839448<span class="elsevierStyleHsp" style=""></span>h after shock diagnosis&#58;LR&#58; 0&#46;9710MLP&#58; 0&#46;9928SVM&#58; 1&#46;0000&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#60;16<span class="elsevierStyleHsp" style=""></span>h after shock diagnosis&#58;LR&#58; 0&#46;5556MLP&#58; 0&#46;4444SVM&#58; 0&#46;722248<span class="elsevierStyleHsp" style=""></span>h after shock diagnosis&#58;LR&#58; 1&#46;0000MLP&#58; 1&#46;000SVM&#58; 1&#46;0000&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&#60;16<span class="elsevierStyleHsp" style=""></span>h after shock diagnosis&#58;LR&#58; 0&#46;9275MLP&#58; 1&#46;0000SVM&#58; 0&#46;956548<span class="elsevierStyleHsp" style=""></span>h after shock diagnosis&#58;LR&#58; 0&#46;9420MLP&#58; 0&#46;9855SVM&#58; 1&#46;0000&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Wang T&#46; et al&#46;&#44; 2018<a class="elsevierStyleCrossRef" href="#bib0715"><span class="elsevierStyleSup">62</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">DBN&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">24 &#40;age&#44; 8 vital signs&#44; 12 laboratory test&#44; GCS&#44; 2 treatments&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Infected ICU-patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;913 &#40;0&#46;906&#8211;0&#46;919&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;825 &#40;0&#46;802&#8211;0&#46;849&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">0&#46;874 &#40;0&#46;802&#8211;0&#46;849&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Garc&#237;a-Gallo et al&#46;&#44; 2018<a class="elsevierStyleCrossRef" href="#bib0740"><span class="elsevierStyleSup">67</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LASSOSGB&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">47 &#40;14 laboratory tests&#44; 9 vital signs&#44; 4 data taken at the time of ICU admission&#44; 14 comorbidities&#44; and 6 organ dysfunction&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ICU patients with sepsis&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">LASSO&#58; 0&#46;792 &#40;0&#46;791&#8211;0&#46;793&#41;SGB&#58; 0&#46;8039 &#40;0&#46;8033&#8211;0&#46;8045&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Chiew et al&#46;&#44; 2019<a class="elsevierStyleCrossRef" href="#bib0725"><span class="elsevierStyleSup">64</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">kNNRFABGTBSVM&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">28 &#40;6 vital signs&#44; 22 HRV parameters&#41;&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ED patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">kNN&#58; 0&#46;06RF&#58; 0&#46;56AB&#58; 0&#46;38GTB&#58; 0&#46;50SVM&#58; 0&#46;63&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr><tr title="table-row"><td class="td-with-role" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t ; entry_with_role_rowhead " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">Perng et al&#46;&#44; 2019<a class="elsevierStyleCrossRef" href="#bib0705"><span class="elsevierStyleSup">60</span></a>&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">RFkNNSVMSoftMaxCombined with three feature extraction methods&#58;CNNAEPCA&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">53 &#40;demographic data&#44; vital signs&#44; laboratory results&#41; obtained during ED stay&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">ED patients&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="left" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">72<span class="elsevierStyleHsp" style=""></span>h mortality &#8211; Best Classifier&#58;CNN<span class="elsevierStyleHsp" style=""></span>&#43;<span class="elsevierStyleHsp" style=""></span>SoftMax&#58; 0&#46;9428 days mortality &#8211; Best Classifier&#58;CNN<span class="elsevierStyleHsp" style=""></span>&#43;<span class="elsevierStyleHsp" style=""></span>SoftMax&#58; 0&#46;92&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td><td class="td" title="\n
                  \t\t\t\t\ttable-entry\n
                  \t\t\t\t  " align="" valign="\n
                  \t\t\t\t\ttop\n
                  \t\t\t\t">&nbsp;\t\t\t\t\t\t\n
                  \t\t\t\t</td></tr></tbody></table>
                  """
              ]
              "imagenFichero" => array:1 [
                0 => "xTab2844746.png"
              ]
            ]
          ]
        ]
        "descripcion" => array:1 [
          "en" => "<p id="spar0080" class="elsevierStyleSimplePara elsevierViewall">Main characteristics of the ML classifiers used for mortality prediction&#46;</p>"
        ]
      ]
      5 => array:5 [
        "identificador" => "upi0005"
        "tipo" => "MULTIMEDIAECOMPONENTE"
        "mostrarFloat" => false
        "mostrarDisplay" => true
        "Ecomponente" => array:2 [
          "fichero" => "mmc1.pdf"
          "ficheroTamanyo" => 63077
        ]
      ]
    ]
    "bibliografia" => array:2 [
      "titulo" => "References"
      "seccion" => array:1 [
        0 => array:2 [
          "identificador" => "bibs0015"
          "bibliografiaReferencia" => array:81 [
            0 => array:3 [
              "identificador" => "bib0410"
              "etiqueta" => "1"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Severe sepsis and septic shock"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:2 [
                            0 => "D&#46;C&#46; Angus"
                            1 => "T&#46; van der Poll"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1056/NEJMra1208623"
                      "Revista" => array:7 [
                        "tituloSerie" => "N&#46; Engl&#46; J&#46; Med&#46;"
                        "fecha" => "2013"
                        "volumen" => "369"
                        "paginaInicial" => "840"
                        "paginaFinal" => "851"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/23984731"
                            "web" => "Medline"
                          ]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S1542356516303044"
                          "estado" => "S300"
                          "issn" => "15423565"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            1 => array:3 [
              "identificador" => "bib0415"
              "etiqueta" => "2"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Use of explicit ICD9-CM codes to identify adult severe sepsis&#58; impacts on epidemiological estimates"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:3 [
                            0 => "C&#46; Bouza"
                            1 => "T&#46; Lopez-Cuadrado"
                            2 => "J&#46;M&#46; Amate-Blanco"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1186/s13054-016-1497-9"
                      "Revista" => array:5 [
                        "tituloSerie" => "Crit&#46; Care"
                        "fecha" => "2016"
                        "volumen" => "20"
                        "paginaInicial" => "313"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/27716355"
                            "web" => "Medline"
                          ]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            2 => array:3 [
              "identificador" => "bib0420"
              "etiqueta" => "3"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Incidence and trends of sepsis in US hospitals using clinical vs claims data&#44; 2009&#8211;2014"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [
                            0 => "C&#46; Rhee"
                            1 => "R&#46; Dantes"
                            2 => "L&#46; Epstein"
                            3 => "D&#46;J&#46; Murphy"
                            4 => "C&#46;W&#46; Seymour"
                            5 => "T&#46;J&#46; Iwashyna"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1001/jama.2017.13836"
                      "Revista" => array:6 [
                        "tituloSerie" => "JAMA"
                        "fecha" => "2017"
                        "volumen" => "318"
                        "paginaInicial" => "1241"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/28903154"
                            "web" => "Medline"
                          ]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S0168827814003924"
                          "estado" => "S300"
                          "issn" => "01688278"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            3 => array:3 [
              "identificador" => "bib0425"
              "etiqueta" => "4"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Sepsis and septic shock"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:4 [
                            0 => "M&#46; Cecconi"
                            1 => "L&#46; Evans"
                            2 => "M&#46; Levy"
                            3 => "A&#46; Rhodes"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/S0140-6736(18)30696-2"
                      "Revista" => array:6 [
                        "tituloSerie" => "Lancet &#40;London&#44; England&#41;"
                        "fecha" => "2018"
                        "volumen" => "392"
                        "paginaInicial" => "75"
                        "paginaFinal" => "87"
                        "itemHostRev" => array:3 [
                          "pii" => "S1542356516303044"
                          "estado" => "S300"
                          "issn" => "15423565"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            4 => array:3 [
              "identificador" => "bib0430"
              "etiqueta" => "5"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [
                            0 => "R&#46;C&#46; Bone"
                            1 => "R&#46;A&#46; Balk"
                            2 => "F&#46;B&#46; Cerra"
                            3 => "R&#46;P&#46; Dellinger"
                            4 => "A&#46;M&#46; Fein"
                            5 => "W&#46;A&#46; Knaus"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1378/chest.101.6.1644"
                      "Revista" => array:6 [
                        "tituloSerie" => "Chest"
                        "fecha" => "1992"
                        "volumen" => "101"
                        "paginaInicial" => "1644"
                        "paginaFinal" => "1655"
                        "itemHostRev" => array:3 [
                          "pii" => "S1542356511003417"
                          "estado" => "S300"
                          "issn" => "15423565"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            5 => array:3 [
              "identificador" => "bib0435"
              "etiqueta" => "6"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "2001 SCCM&#47;ESICM&#47;ACCP&#47;ATS&#47;SIS international sepsis definitions conference"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [
                            0 => "M&#46;M&#46; Levy"
                            1 => "M&#46;P&#46; Fink"
                            2 => "J&#46;C&#46; Marshall"
                            3 => "E&#46; Abraham"
                            4 => "D&#46; Angus"
                            5 => "D&#46; Cook"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/01.CCM.0000050454.01978.3B"
                      "Revista" => array:7 [
                        "tituloSerie" => "Crit&#46; Care Med&#46;"
                        "fecha" => "2003"
                        "volumen" => "31"
                        "paginaInicial" => "1250"
                        "paginaFinal" => "1256"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/12682500"
                            "web" => "Medline"
                          ]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S2468125319300044"
                          "estado" => "S300"
                          "issn" => "24681253"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            6 => array:3 [
              "identificador" => "bib0440"
              "etiqueta" => "7"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "The third international consensus definitions for sepsis and septic shock &#40;sepsis-3&#41;"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [
                            0 => "M&#46; Singer"
                            1 => "C&#46;S&#46; Deutschman"
                            2 => "C&#46;W&#46; Seymour"
                            3 => "M&#46; Shankar-Hari"
                            4 => "D&#46; Annane"
                            5 => "M&#46; Bauer"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1001/jama.2016.0287"
                      "Revista" => array:6 [
                        "tituloSerie" => "JAMA"
                        "fecha" => "2016"
                        "volumen" => "315"
                        "paginaInicial" => "801"
                        "paginaFinal" => "810"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/26903338"
                            "web" => "Medline"
                          ]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            7 => array:3 [
              "identificador" => "bib0445"
              "etiqueta" => "8"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Assessment of global incidence and mortality of hospital-treated sepsis current estimates and limitations"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [
                            0 => "C&#46; Fleischmann"
                            1 => "A&#46; Scherag"
                            2 => "N&#46;K&#46;J&#46; Adhikari"
                            3 => "C&#46;S&#46; Hartog"
                            4 => "T&#46; Tsaganos"
                            5 => "P&#46; Schlattmann"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1164/rccm.201504-0781OC"
                      "Revista" => array:6 [
                        "tituloSerie" => "Am&#46; J&#46; Respir&#46; Crit&#46; Care Med&#46;"
                        "fecha" => "2016"
                        "volumen" => "193"
                        "paginaInicial" => "259"
                        "paginaFinal" => "272"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/26414292"
                            "web" => "Medline"
                          ]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            8 => array:3 [
              "identificador" => "bib0450"
              "etiqueta" => "9"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Sepsis incidence and outcome&#58; contrasting the intensive care unit with the hospital ward"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [
                            0 => "A&#46; Esteban"
                            1 => "F&#46; Frutos-Vivar"
                            2 => "N&#46;D&#46; Ferguson"
                            3 => "O&#46; Pe&#241;uelas"
                            4 => "Lorente J&#193;"
                            5 => "F&#46; Gordo"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/01.CCM.0000260960.94300.DE"
                      "Revista" => array:6 [
                        "tituloSerie" => "Crit&#46; Care Med&#46;"
                        "fecha" => "2007"
                        "volumen" => "35"
                        "paginaInicial" => "1284"
                        "paginaFinal" => "1289"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/17414725"
                            "web" => "Medline"
                          ]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            9 => array:3 [
              "identificador" => "bib0455"
              "etiqueta" => "10"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Epidemiology and recent trends of severe sepsis in Spain&#58; a nationwide population-based analysis &#40;2006&#8211;2011&#41;"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:4 [
                            0 => "C&#46; Bouza"
                            1 => "T&#46; L&#243;pez-Cuadrado"
                            2 => "Z&#46; Saz-Parkinson"
                            3 => "J&#46;M&#46; Amate-Blanco"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1186/s12879-014-0717-7"
                      "Revista" => array:5 [
                        "tituloSerie" => "BMC Infect&#46; Dis&#46;"
                        "fecha" => "2014"
                        "volumen" => "14"
                        "paginaInicial" => "3863"
                        "itemHostRev" => array:3 [
                          "pii" => "S0016510719324290"
                          "estado" => "S300"
                          "issn" => "00165107"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            10 => array:3 [
              "identificador" => "bib0460"
              "etiqueta" => "11"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Epidemiology of severe sepsis in the United States&#58; analysis of incidence&#44; outcome&#44; and associated costs of care"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [
                            0 => "D&#46;C&#46; Angus"
                            1 => "W&#46;T&#46; Linde-Zwirble"
                            2 => "J&#46; Lidicker"
                            3 => "G&#46; Clermont"
                            4 => "J&#46; Carcillo"
                            5 => "M&#46;R&#46; Pinsky"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/00003246-200107000-00002"
                      "Revista" => array:7 [
                        "tituloSerie" => "Crit&#46; Care Med&#46;"
                        "fecha" => "2001"
                        "volumen" => "29"
                        "paginaInicial" => "1303"
                        "paginaFinal" => "1310"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/11445675"
                            "web" => "Medline"
                          ]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S0016508510008620"
                          "estado" => "S300"
                          "issn" => "00165085"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            11 => array:3 [
              "identificador" => "bib0465"
              "etiqueta" => "12"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "The Epidemiology of Sepsis in the United States from 1979 through 2000"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:4 [
                            0 => "G&#46;S&#46; Martin"
                            1 => "D&#46;M&#46; Mannino"
                            2 => "S&#46; Eaton"
                            3 => "M&#46; Moss"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1056/NEJMoa022139"
                      "Revista" => array:7 [
                        "tituloSerie" => "N&#46; Engl&#46; J&#46; Med&#46;"
                        "fecha" => "2003"
                        "volumen" => "348"
                        "paginaInicial" => "1546"
                        "paginaFinal" => "1554"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/12700374"
                            "web" => "Medline"
                          ]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S1542356507011020"
                          "estado" => "S300"
                          "issn" => "15423565"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            12 => array:3 [
              "identificador" => "bib0470"
              "etiqueta" => "13"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Cost-effectiveness of the Surviving Sepsis Campaign protocol for severe sepsis&#58; a prospective nation-wide study in Spain"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [
                            0 => "D&#46; Suarez"
                            1 => "R&#46; Ferrer"
                            2 => "A&#46; Artigas"
                            3 => "I&#46; Azkarate"
                            4 => "J&#46; Garnacho-Montero"
                            5 => "G&#46; Gom&#224;"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1007/s00134-010-2102-3"
                      "Revista" => array:6 [
                        "tituloSerie" => "Intensive Care Med&#46;"
                        "fecha" => "2011"
                        "volumen" => "37"
                        "paginaInicial" => "444"
                        "paginaFinal" => "452"
                        "itemHostRev" => array:3 [
                          "pii" => "S1542356516303135"
                          "estado" => "S300"
                          "issn" => "15423565"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            13 => array:3 [
              "identificador" => "bib0475"
              "etiqueta" => "14"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Hospitalizations&#44; costs&#44; and outcomes of severe sepsis in the United States 2003 to 2007"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [
                            0 => "T&#46; Lagu"
                            1 => "M&#46;B&#46; Rothberg"
                            2 => "M&#46;-S&#46; Shieh"
                            3 => "P&#46;S&#46; Pekow"
                            4 => "J&#46;S&#46; Steingrub"
                            5 => "P&#46;K&#46; Lindenauer"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/CCM.0b013e318232db65"
                      "Revista" => array:7 [
                        "tituloSerie" => "Crit&#46; Care Med&#46;"
                        "fecha" => "2012"
                        "volumen" => "40"
                        "paginaInicial" => "754"
                        "paginaFinal" => "761"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/21963582"
                            "web" => "Medline"
                          ]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S0959804916322869"
                          "estado" => "S300"
                          "issn" => "09598049"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            14 => array:3 [
              "identificador" => "bib0480"
              "etiqueta" => "15"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Surviving Sepsis Campaign&#58; International Guidelines for Management of Sepsis and Septic Shock&#58; 2016"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [
                            0 => "A&#46; Rhodes"
                            1 => "L&#46;E&#46; Evans"
                            2 => "W&#46; Alhazzani"
                            3 => "M&#46;M&#46; Levy"
                            4 => "M&#46; Antonelli"
                            5 => "R&#46; Ferrer"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1007/s00134-017-4683-6"
                      "Revista" => array:7 [
                        "tituloSerie" => "Intensive Care Med&#46;"
                        "fecha" => "2017"
                        "volumen" => "43"
                        "paginaInicial" => "304"
                        "paginaFinal" => "377"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/28101605"
                            "web" => "Medline"
                          ]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S2468125321000625"
                          "estado" => "S300"
                          "issn" => "24681253"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            15 => array:3 [
              "identificador" => "bib0485"
              "etiqueta" => "16"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "C&#243;digo sepsis&#44; documento de consenso&#58; recomendaciones&#46; Madrid&#58; IMC&#46; &#91;Internet&#93;"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:5 [
                            0 => "M&#46; Borges-S&#225;"
                            1 => "F&#46; Candel-Gonz&#225;lez"
                            2 => "R&#46; Ferrer-Roca"
                            3 => "P&#46; Cort&#233;s-Vidal"
                            4 => "R&#46; Zaragoza-Crespo"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:1 [
                      "Libro" => array:3 [
                        "fecha" => "2014"
                        "editorial" => "Espa&#241;a"
                        "editorialLocalizacion" => "Madrid"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            16 => array:3 [
              "identificador" => "bib0490"
              "etiqueta" => "17"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Impact of adequate empirical antibiotic therapy on the outcome of patients admitted to the intensive care unit with sepsis"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [
                            0 => "J&#46; Garnacho-Montero"
                            1 => "J&#46;L&#46; Garcia-Garmendia"
                            2 => "A&#46; Barrero-Almodovar"
                            3 => "F&#46;J&#46; Jimenez-Jimenez"
                            4 => "C&#46; Perez-Paredes"
                            5 => "C&#46; Ortiz-Leyba"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/01.CCM.0000098031.24329.10"
                      "Revista" => array:6 [
                        "tituloSerie" => "Crit&#46; Care Med&#46;"
                        "fecha" => "2003"
                        "volumen" => "31"
                        "paginaInicial" => "2742"
                        "paginaFinal" => "2751"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/14668610"
                            "web" => "Medline"
                          ]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            17 => array:3 [
              "identificador" => "bib0495"
              "etiqueta" => "18"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [
                            0 => "A&#46; Kumar"
                            1 => "D&#46; Roberts"
                            2 => "K&#46;E&#46; Wood"
                            3 => "B&#46; Light"
                            4 => "J&#46;E&#46; Parrillo"
                            5 => "S&#46; Sharma"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/01.CCM.0000217961.75225.E9"
                      "Revista" => array:6 [
                        "tituloSerie" => "Crit&#46; Care Med&#46;"
                        "fecha" => "2006"
                        "volumen" => "34"
                        "paginaInicial" => "1589"
                        "paginaFinal" => "1596"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/16625125"
                            "web" => "Medline"
                          ]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            18 => array:3 [
              "identificador" => "bib0500"
              "etiqueta" => "19"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Fluid administration for acute circulatory dysfunction using basic monitoring&#58; narrative review and expert panel recommendations from an ESICM task force"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [
                            0 => "M&#46; Cecconi"
                            1 => "G&#46; Hernandez"
                            2 => "M&#46; Dunser"
                            3 => "M&#46; Antonelli"
                            4 => "T&#46; Baker"
                             …1
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1007/s00134-018-5415-2"
                      "Revista" => array:6 [
                        "tituloSerie" => "Intensive Care Med&#46;"
                        "fecha" => "2019"
                        "volumen" => "45"
                        "paginaInicial" => "21"
                        "paginaFinal" => "32"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            19 => array:3 [
              "identificador" => "bib0505"
              "etiqueta" => "20"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Impact of source control in patients with severe sepsis and septic shock"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/CCM.0000000000002011"
                      "Revista" => array:6 [
                        "tituloSerie" => "Crit&#46; Care Med&#46;"
                        "fecha" => "2017"
                        "volumen" => "45"
                        "paginaInicial" => "11"
                        "paginaFinal" => "19"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            20 => array:3 [
              "identificador" => "bib0510"
              "etiqueta" => "21"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "The surviving sepsis campaign bundle&#58; 2018 update"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:3 [ …3]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1007/s00134-018-5085-0"
                      "Revista" => array:7 [
                        "tituloSerie" => "Intensive Care Med&#46;"
                        "fecha" => "2018"
                        "volumen" => "44"
                        "paginaInicial" => "925"
                        "paginaFinal" => "928"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S016882781730185X"
                          "estado" => "S300"
                          "issn" => "01688278"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            21 => array:3 [
              "identificador" => "bib0515"
              "etiqueta" => "22"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Surviving Sepsis Campaign&#58; association between performance metrics and outcomes in a 7&#46;5-year study"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/CCM.0000000000000723"
                      "Revista" => array:7 [
                        "tituloSerie" => "Crit&#46; Care Med&#46;"
                        "fecha" => "2015"
                        "volumen" => "43"
                        "paginaInicial" => "3"
                        "paginaFinal" => "12"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S2468125321000571"
                          "estado" => "S300"
                          "issn" => "24681253"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            22 => array:3 [
              "identificador" => "bib0520"
              "etiqueta" => "23"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Effect of performance improvement programs on compliance with sepsis bundles and mortality&#58; a systematic review and meta-analysis of observational studies Efron PA editor"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1371/journal.pone.0125827"
                      "Revista" => array:5 [
                        "tituloSerie" => "PLoS One"
                        "fecha" => "2015"
                        "volumen" => "10"
                        "paginaInicial" => "e0125827"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            23 => array:3 [
              "identificador" => "bib0525"
              "etiqueta" => "24"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "The surviving sepsis campaign bundles and outcome&#58; results from the international multicentre prevalence study on sepsis &#40;the IMPreSS study&#41;"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1007/s00134-015-3906-y"
                      "Revista" => array:6 [
                        "tituloSerie" => "Intensive Care Med&#46;"
                        "fecha" => "2015"
                        "volumen" => "41"
                        "paginaInicial" => "1620"
                        "paginaFinal" => "1628"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            24 => array:3 [
              "identificador" => "bib0530"
              "etiqueta" => "25"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Improvement in process of care and outcome after a multicenter severe sepsis educational program in Spain"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1001/jama.299.19.2294"
                      "Revista" => array:6 [
                        "tituloSerie" => "JAMA"
                        "fecha" => "2008"
                        "volumen" => "299"
                        "paginaInicial" => "2294"
                        "paginaFinal" => "2303"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            25 => array:3 [
              "identificador" => "bib0535"
              "etiqueta" => "26"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Impacto de la implantaci&#243;n de un C&#243;digo Sepsis intrahospitalario en la prescripci&#243;n de antibi&#243;ticos y los resultados cl&#237;nicos en una unidad de cuidados intensivos"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/j.medin.2016.08.001"
                      "Revista" => array:6 [
                        "tituloSerie" => "Med Intensiva"
                        "fecha" => "2017"
                        "volumen" => "41"
                        "paginaInicial" => "12"
                        "paginaFinal" => "20"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            26 => array:3 [
              "identificador" => "bib0540"
              "etiqueta" => "27"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Time to treatment and mortality during mandated emergency care for sepsis"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1056/NEJMoa1703058"
                      "Revista" => array:6 [
                        "tituloSerie" => "N&#46; Engl&#46; J&#46; Med&#46;"
                        "fecha" => "2017"
                        "volumen" => "376"
                        "paginaInicial" => "2235"
                        "paginaFinal" => "2244"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            27 => array:3 [
              "identificador" => "bib0545"
              "etiqueta" => "28"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Automated detection of sepsis using electronic medical record data"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:1 [ …1]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/JHQ.0000000000000066"
                      "Revista" => array:6 [
                        "tituloSerie" => "J&#46; Healthc&#46; Qual&#46;"
                        "fecha" => "2017"
                        "volumen" => "39"
                        "paginaInicial" => "322"
                        "paginaFinal" => "333"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            28 => array:3 [
              "identificador" => "bib0550"
              "etiqueta" => "29"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Automated monitoring compared to standard care for the early detection of sepsis in critically ill patients"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1002/14651858.CD012404.pub2"
                      "Revista" => array:5 [
                        "tituloSerie" => "Cochrane Database Syst Rev"
                        "fecha" => "2018"
                        "volumen" => "6"
                        "paginaInicial" => "1465"
                        "paginaFinal" => "1858"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            29 => array:3 [
              "identificador" => "bib0555"
              "etiqueta" => "30"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Computer versus paper system for recognition and management of sepsis in surgical intensive care"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/TA.0000000000000121"
                      "Revista" => array:6 [
                        "tituloSerie" => "J Trauma Acute Care Surg"
                        "fecha" => "2014"
                        "volumen" => "76"
                        "paginaInicial" => "311"
                        "paginaFinal" => "319"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            30 => array:3 [
              "identificador" => "bib0560"
              "etiqueta" => "31"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Predicting the future &#8211; big data&#44; machine learning&#44; and clinical medicine"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1056/NEJMp1606181"
                      "Revista" => array:7 [
                        "tituloSerie" => "N&#46; Engl&#46; J&#46; Med&#46;"
                        "fecha" => "2016"
                        "volumen" => "375"
                        "paginaInicial" => "1216"
                        "paginaFinal" => "1219"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S0735109708026260"
                          "estado" => "S300"
                          "issn" => "07351097"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            31 => array:3 [
              "identificador" => "bib0565"
              "etiqueta" => "32"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "A survey on deep learning in medical image analysis"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/J.MEDIA.2017.07.005"
                      "Revista" => array:7 [
                        "tituloSerie" => "Med&#46; Image Anal&#46;"
                        "fecha" => "2017"
                        "volumen" => "42"
                        "paginaInicial" => "60"
                        "paginaFinal" => "88"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S0016508513000723"
                          "estado" => "S300"
                          "issn" => "00165085"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            32 => array:3 [
              "identificador" => "bib0570"
              "etiqueta" => "33"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Machine learning in cardiovascular medicine&#58; are we there yet&#63;"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:5 [ …5]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1136/heartjnl-2017-311198"
                      "Revista" => array:6 [
                        "tituloSerie" => "Heart"
                        "fecha" => "2018"
                        "volumen" => "104"
                        "paginaInicial" => "1156"
                        "paginaFinal" => "1164"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            33 => array:3 [
              "identificador" => "bib0575"
              "etiqueta" => "34"
              "referencia" => array:1 [
                0 => array:1 [
                  "referenciaCompleta" => "Organizing Committee of the Madrid Critical Care&#44; D&#46;&#44; et al&#46; Big data and machine learning in critical care&#58; Opportunities for collaborative research&#46; Med Intensiva&#46; 2019&#59;43&#58;52&#8211;7&#46; <a target="_blank" href="doi:10.1016/j.medin.2018.06.002">doi&#58;10&#46;1016&#47;j&#46;medin&#46;2018&#46;06&#46;002</a>&#46;"
                ]
              ]
            ]
            34 => array:3 [
              "identificador" => "bib0580"
              "etiqueta" => "35"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Machine learning in neuro-oncology&#58; can data analysis from 5346 patients change decision-making paradigms&#63;"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/j.wneu.2019.01.046"
                      "Revista" => array:5 [
                        "tituloSerie" => "World Neurosurg"
                        "fecha" => "2019"
                        "volumen" => "124"
                        "paginaInicial" => "287"
                        "paginaFinal" => "294"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            35 => array:3 [
              "identificador" => "bib0585"
              "etiqueta" => "36"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "A new effective machine learning framework for sepsis diagnosis"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1109/ACCESS.2018.2867728"
                      "Revista" => array:6 [
                        "tituloSerie" => "IEEE Access"
                        "fecha" => "2018"
                        "volumen" => "6"
                        "paginaInicial" => "48300"
                        "paginaFinal" => "48310"
                        "itemHostRev" => array:3 [
                          "pii" => "S2468125321000546"
                          "estado" => "S300"
                          "issn" => "24681253"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            36 => array:3 [
              "identificador" => "bib0590"
              "etiqueta" => "37"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Learning representations for the early detection of sepsis with deep neural networks"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/j.compbiomed.2017.08.015"
                      "Revista" => array:6 [
                        "tituloSerie" => "Comput&#46; Biol&#46; Med&#46;"
                        "fecha" => "2017"
                        "volumen" => "89"
                        "paginaInicial" => "248"
                        "paginaFinal" => "255"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            37 => array:3 [
              "identificador" => "bib0595"
              "etiqueta" => "38"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "An improved multi-output gaussian process RNN with real-time validation for early sepsis detection"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:1 [
                      "Revista" => array:2 [
                        "tituloSerie" => "Arxiv&#91;Preprint&#93;"
                        "fecha" => "2017"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            38 => array:3 [
              "identificador" => "bib0600"
              "etiqueta" => "39"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Predict sepsis level in intensive medicine &#8211; data mining approach"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1007/978-3-642-36981-0_19"
                      "Libro" => array:6 [
                        "titulo" => "Advances in information systems and technologies"
                        "fecha" => "2013"
                        "paginaInicial" => "201"
                        "paginaFinal" => "211"
                        "editorial" => "Heidelberg"
                        "editorialLocalizacion" => "Springer&#44; Berlin"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            39 => array:3 [
              "identificador" => "bib0605"
              "etiqueta" => "40"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Early detection of sepsis in the emergency department using Dynamic Bayesian Networks"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:1 [
                      "Libro" => array:4 [
                        "titulo" => "AMIA&#46; Annual symposium proceedings AMIA symposium"
                        "fecha" => "2012"
                        "paginaInicial" => "653"
                        "paginaFinal" => "662"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            40 => array:3 [
              "identificador" => "bib0610"
              "etiqueta" => "41"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Prediction of sepsis in the intensive care unit with minimal electronic health record data&#58; a machine learning approach"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.2196/medinform.5909"
                      "Revista" => array:5 [
                        "tituloSerie" => "JMIR Med Inform"
                        "fecha" => "2016"
                        "volumen" => "4"
                        "paginaInicial" => "e28"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            41 => array:3 [
              "identificador" => "bib0615"
              "etiqueta" => "42"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Multicentre validation of a sepsis prediction algorithm using only vital sign data in the emergency department&#44; general ward and ICU"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1136/bmjopen-2017-017833"
                      "Revista" => array:6 [
                        "tituloSerie" => "BMJ Open"
                        "fecha" => "2018"
                        "volumen" => "8"
                        "paginaInicial" => "e017833"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S0016508510008620"
                          "estado" => "S300"
                          "issn" => "00165085"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            42 => array:3 [
              "identificador" => "bib0620"
              "etiqueta" => "43"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Development and evaluation of a machine learning model for the early identification of patients at risk for sepsis"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:5 [ …5]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/j.annemergmed.2018.11.036"
                      "Revista" => array:6 [
                        "tituloSerie" => "Ann&#46; Emerg&#46; Med&#46;"
                        "fecha" => "2019"
                        "volumen" => "73"
                        "paginaInicial" => "334"
                        "paginaFinal" => "344"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            43 => array:3 [
              "identificador" => "bib0625"
              "etiqueta" => "44"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Machine-learning-based laboratory developed test for the diagnosis of sepsis in high-risk patients"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:4 [ …4]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.3390/diagnostics9010020"
                      "Revista" => array:4 [
                        "tituloSerie" => "Diagnostics"
                        "fecha" => "2019"
                        "volumen" => "9"
                        "paginaInicial" => "20"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            44 => array:3 [
              "identificador" => "bib0630"
              "etiqueta" => "45"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Evaluation of a machine learning algorithm for up to 48-hour advance prediction of sepsis using six vital signs"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/j.compbiomed.2019.04.027"
                      "Revista" => array:5 [
                        "tituloSerie" => "Comput&#46; Biol&#46; Med&#46;"
                        "fecha" => "2019"
                        "volumen" => "109"
                        "paginaInicial" => "79"
                        "paginaFinal" => "84"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            45 => array:3 [
              "identificador" => "bib0635"
              "etiqueta" => "46"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Development and validation of an automated sepsis risk assessment system"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:4 [ …4]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1002/nur.21734"
                      "Revista" => array:6 [
                        "tituloSerie" => "Res&#46; Nurs&#46; Health"
                        "fecha" => "2016"
                        "volumen" => "39"
                        "paginaInicial" => "317"
                        "paginaFinal" => "327"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            46 => array:3 [
              "identificador" => "bib0640"
              "etiqueta" => "47"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Development and external validation of an automated computer-aided risk score for predicting sepsis in emergency medical admissions using the patient&#39;s first electronically recorded vital signs and blood test results"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/CCM.0000000000002967"
                      "Revista" => array:6 [
                        "tituloSerie" => "Crit&#46; Care Med&#46;"
                        "fecha" => "2018"
                        "volumen" => "46"
                        "paginaInicial" => "612"
                        "paginaFinal" => "618"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            47 => array:3 [
              "identificador" => "bib0645"
              "etiqueta" => "48"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Creating an automated trigger for sepsis clinical decision support at emergency department triage using machine learning Groza T editor"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1371/journal.pone.0174708"
                      "Revista" => array:6 [
                        "tituloSerie" => "PLOS ONE"
                        "fecha" => "2017"
                        "volumen" => "12"
                        "paginaInicial" => "e0174708"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S2468125320302387"
                          "estado" => "S300"
                          "issn" => "24681253"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            48 => array:3 [
              "identificador" => "bib0650"
              "etiqueta" => "49"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Non-invasive classification of severe sepsis and systemic inflammatory response syndrome using a nonlinear support vector machine&#58; a preliminary study"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1088/0967-3334/31/6/004"
                      "Revista" => array:6 [
                        "tituloSerie" => "Physiol&#46; Meas&#46;"
                        "fecha" => "2010"
                        "volumen" => "31"
                        "paginaInicial" => "775"
                        "paginaFinal" => "793"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            49 => array:3 [
              "identificador" => "bib0655"
              "etiqueta" => "50"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Natural language processing of electronic medical records can identify sepsis following orthopedic surgery"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:5 [ …5]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/J.SPINEE.2018.06.068"
                      "Revista" => array:4 [
                        "tituloSerie" => "Spine J"
                        "fecha" => "2018"
                        "volumen" => "18"
                        "paginaInicial" => "29"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            50 => array:3 [
              "identificador" => "bib0660"
              "etiqueta" => "51"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Leveraging implicit expert knowledge for non-circular machine learning in sepsis prediction"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:5 [ …5]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/j.artmed.2019.101725"
                      "Revista" => array:5 [
                        "tituloSerie" => "Artif&#46; Intell&#46; Med&#46;"
                        "fecha" => "2019"
                        "volumen" => "100"
                        "paginaInicial" => "101725"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            51 => array:3 [
              "identificador" => "bib0665"
              "etiqueta" => "52"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Predicting sepsis with a recurrent neural network using the MIMIC III database"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:4 [ …4]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/j.compbiomed.2019.103395"
                      "Revista" => array:5 [
                        "tituloSerie" => "Comput&#46; Biol&#46; Med&#46;"
                        "fecha" => "2019"
                        "volumen" => "113"
                        "paginaInicial" => "103395"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            52 => array:3 [
              "identificador" => "bib0670"
              "etiqueta" => "53"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "An attention based deep learning model of clinical events in the intensive care unit"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1371/journal.pone.0211057"
                      "Revista" => array:5 [
                        "tituloSerie" => "PLOS ONE"
                        "fecha" => "2019"
                        "volumen" => "14"
                        "paginaInicial" => "e0211057"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            53 => array:3 [
              "identificador" => "bib0675"
              "etiqueta" => "54"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "LiSep LSTM&#58; a machine learning algorithm for early detection of septic shock"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:4 [ …4]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1038/s41598-019-51219-4"
                      "Revista" => array:7 [
                        "tituloSerie" => "Sci Rep"
                        "fecha" => "2019"
                        "volumen" => "9"
                        "paginaInicial" => "1"
                        "paginaFinal" => "8"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S2468125319300044"
                          "estado" => "S300"
                          "issn" => "24681253"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            54 => array:3 [
              "identificador" => "bib0680"
              "etiqueta" => "55"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Predictive Models of Sepsis in Adult ICU Patients&#46; In&#58; 2018 IEEE International Conference on Healthcare Informatics &#40;ICHI&#41;"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1109/ICHI.2018.00068"
                      "Libro" => array:5 [
                        "fecha" => "2018"
                        "paginaInicial" => "390"
                        "paginaFinal" => "391"
                        "editorial" => "IEEE"
                        "editorialLocalizacion" => "New York&#44; NY&#44; USA"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            55 => array:3 [
              "identificador" => "bib0685"
              "etiqueta" => "56"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "A machine learning algorithm to predict severe sepsis and septic shock&#58; development implementation&#44; and impact on clinical practice"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/CCM.0000000000003891"
                      "Revista" => array:4 [
                        "tituloSerie" => "Crit&#46; Care Med&#46;"
                        "fecha" => "2019"
                        "volumen" => "47"
                        "paginaInicial" => "e20"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            56 => array:3 [
              "identificador" => "bib0690"
              "etiqueta" => "57"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "A minimal set of physiomarkers in continuous high frequency data streams predict adult sepsis onset earlier"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/j.ijmedinf.2018.12.002"
                      "Revista" => array:6 [
                        "tituloSerie" => "Int J Med Inform"
                        "fecha" => "2019"
                        "volumen" => "122"
                        "paginaInicial" => "55"
                        "paginaFinal" => "62"
                        "itemHostRev" => array:3 [
                          "pii" => "S0002870320303586"
                          "estado" => "S300"
                          "issn" => "00028703"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            57 => array:3 [
              "identificador" => "bib0695"
              "etiqueta" => "58"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "An interpretable machine learning model for accurate prediction of sepsis in the ICU"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/CCM.0000000000002936"
                      "Revista" => array:6 [
                        "tituloSerie" => "Crit&#46; Care Med&#46;"
                        "fecha" => "2018"
                        "volumen" => "46"
                        "paginaInicial" => "547"
                        "paginaFinal" => "553"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            58 => array:3 [
              "identificador" => "bib0700"
              "etiqueta" => "59"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Using peptidomics and machine learning techniques to predict mortality of patients with septic shock &#91;Internet&#93;"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:1 [ …1]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:1 [
                      "Libro" => array:2 [
                        "fecha" => "2018"
                        "editorial" => "Universitat Polit&#232;cnica de Catalunya"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            59 => array:3 [
              "identificador" => "bib0705"
              "etiqueta" => "60"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Mortality prediction of septic patients in the emergency department based on machine learning"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.3390/jcm8111906"
                      "Revista" => array:4 [
                        "tituloSerie" => "J Clin Med"
                        "fecha" => "2019"
                        "volumen" => "8"
                        "paginaInicial" => "1906"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            60 => array:3 [
              "identificador" => "bib0710"
              "etiqueta" => "61"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "From vital signs to clinical outcomes for patients with sepsis&#58; a machine learning basis for a clinical decision support system"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1136/amiajnl-2013-001815"
                      "Revista" => array:6 [
                        "tituloSerie" => "J Am Med Informatics Assoc"
                        "fecha" => "2014"
                        "volumen" => "21"
                        "paginaInicial" => "315"
                        "paginaFinal" => "325"
                        "itemHostRev" => array:3 [
                          "pii" => "S2468125320303940"
                          "estado" => "S300"
                          "issn" => "24681253"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            61 => array:3 [
              "identificador" => "bib0715"
              "etiqueta" => "62"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Semantically enhanced dynamic bayesian network for detecting sepsis mortality risk in ICU patients with infection"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:1 [
                      "Revista" => array:2 [
                        "tituloSerie" => "Arxiv &#91;Preprint&#93;"
                        "fecha" => "2018"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            62 => array:3 [
              "identificador" => "bib0720"
              "etiqueta" => "63"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Prediction of in-hospital mortality in emergency department patients with sepsis&#58; a local big data-driven machine learning approach"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1111/acem.12876"
                      "Revista" => array:6 [
                        "tituloSerie" => "Acad&#46; Emerg&#46; Med&#46;"
                        "fecha" => "2016"
                        "volumen" => "23"
                        "paginaInicial" => "269"
                        "paginaFinal" => "278"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            63 => array:3 [
              "identificador" => "bib0725"
              "etiqueta" => "64"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Heart rate variability based machine learning models for risk prediction of suspected sepsis patients in the emergency department"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1097/MD.0000000000014197"
                      "Revista" => array:5 [
                        "tituloSerie" => "Med &#40;United States&#41;"
                        "fecha" => "2019"
                        "volumen" => "98"
                        "paginaInicial" => "e14197"
                        "itemHostRev" => array:3 [
                          "pii" => "S2468125321000546"
                          "estado" => "S300"
                          "issn" => "24681253"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            64 => array:3 [
              "identificador" => "bib0730"
              "etiqueta" => "65"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Severe sepsis mortality prediction with logistic regression over latent factors"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:4 [ …4]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/J.ESWA.2011.08.054"
                      "Revista" => array:6 [
                        "tituloSerie" => "Expert Syst Appl"
                        "fecha" => "2012"
                        "volumen" => "39"
                        "paginaInicial" => "1937"
                        "paginaFinal" => "1943"
                        "itemHostRev" => array:3 [
                          "pii" => "S2468125321000625"
                          "estado" => "S300"
                          "issn" => "24681253"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            65 => array:3 [
              "identificador" => "bib0735"
              "etiqueta" => "66"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "From data to optimal decision making&#58; a data-driven probabilistic machine learning approach to decision support for patients with sepsis"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:3 [ …3]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.2196/medinform.3445"
                      "Revista" => array:4 [
                        "tituloSerie" => "JMIR Med Inform"
                        "fecha" => "2015"
                        "volumen" => "3"
                        "paginaInicial" => "e11"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            66 => array:3 [
              "identificador" => "bib0740"
              "etiqueta" => "67"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "A machine learning-based model for 1-year mortality prediction in patients admitted to an Intensive Care Unit with a diagnosis of sepsis"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:4 [ …4]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/j.medin.2018.07.016"
                      "Revista" => array:4 [
                        "tituloSerie" => "Med Intensiva"
                        "fecha" => "2018"
                        "paginaInicial" => "30245"
                        "paginaFinal" => "30246"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            67 => array:3 [
              "identificador" => "bib0745"
              "etiqueta" => "68"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Introduction to machine learning &#91;Internet&#93;"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:1 [ …1]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:1 [
                      "Libro" => array:2 [
                        "fecha" => "2010"
                        "editorial" => "The MIT Press"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            68 => array:3 [
              "identificador" => "bib0750"
              "etiqueta" => "69"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Early diagnosis and prediction of sepsis shock by combining static and dynamic information using convolutional-LSTM"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1109/ICHI. 2018.00032"
                      "Libro" => array:6 [
                        "titulo" => "2018 IEEE International Conference on Healthcare Informatics &#40;ICHI&#41;"
                        "fecha" => "2018"
                        "paginaInicial" => "219"
                        "paginaFinal" => "228"
                        "editorial" => "IEEE"
                        "editorialLocalizacion" => "New York&#44; NY&#44; USA"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            69 => array:3 [
              "identificador" => "bib0755"
              "etiqueta" => "70"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Data-driven discovery of a novel sepsis pre-shock state predicts impending septic shock in the ICU"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:1 [
                      "Revista" => array:4 [
                        "tituloSerie" => "Sci Rep"
                        "fecha" => "2019"
                        "volumen" => "9"
                        "paginaInicial" => "6145"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            70 => array:3 [
              "identificador" => "bib0760"
              "etiqueta" => "71"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "MIMIC-III&#44; a freely accessible critical care database"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1038/sdata.2016.35"
                      "Revista" => array:5 [
                        "tituloSerie" => "Sci Data"
                        "fecha" => "2016"
                        "volumen" => "3"
                        "paginaInicial" => "160035"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            71 => array:3 [
              "identificador" => "bib0765"
              "etiqueta" => "72"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "The Artificial Intelligence Clinician learns optimal treatment strategies for sepsis in intensive care"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:5 [ …5]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1038/s41591-018-0213-5"
                      "Revista" => array:5 [
                        "tituloSerie" => "Nat&#46; Med&#46;"
                        "fecha" => "2018"
                        "volumen" => "24"
                        "paginaInicial" => "1716"
                        "paginaFinal" => "1720"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            72 => array:3 [
              "identificador" => "bib0770"
              "etiqueta" => "73"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "The eICU collaborative research database&#44; a freely available multi-center database for critical care research"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1038/sdata.2018.178"
                      "Revista" => array:6 [
                        "tituloSerie" => "Sci Data"
                        "fecha" => "2018"
                        "volumen" => "5"
                        "paginaInicial" => "1"
                        "paginaFinal" => "13"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            73 => array:3 [
              "identificador" => "bib0775"
              "etiqueta" => "74"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Reducing patient mortality length of stay and readmissions through machine learning-based sepsis prediction in the emergency department&#44; intensive care unit and hospital floor units"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1136/bmjoq-2017-000158"
                      "Revista" => array:5 [
                        "tituloSerie" => "BMJ Open Qual"
                        "fecha" => "2017"
                        "volumen" => "6"
                        "paginaInicial" => "e000158"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            74 => array:3 [
              "identificador" => "bib0780"
              "etiqueta" => "75"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Effect of a machine learning-based severe sepsis prediction algorithm on patient survival and hospital length of stay&#58; a randomised clinical trial"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:5 [ …5]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1136/bmjresp-2017-000234"
                      "Revista" => array:6 [
                        "tituloSerie" => "BMJ Open Respir Res"
                        "fecha" => "2017"
                        "volumen" => "4"
                        "paginaInicial" => "e000234"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S2468125319303462"
                          "estado" => "S300"
                          "issn" => "24681253"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            75 => array:3 [
              "identificador" => "bib0785"
              "etiqueta" => "76"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Big data and machine learning in health care&#46; Vol&#46; 319&#44; JAMA &#8211; Journal of the American Medical Association"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:1 [
                      "Libro" => array:4 [
                        "fecha" => "2018"
                        "paginaInicial" => "1317"
                        "paginaFinal" => "1318"
                        "editorial" => "American Medical Association"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            76 => array:3 [
              "identificador" => "bib0790"
              "etiqueta" => "77"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Potential biases in machine learning algorithms using electronic health record data"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:4 [ …4]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1001/jamainternmed.2018.3763"
                      "Revista" => array:4 [
                        "tituloSerie" => "JAMA Intern Med"
                        "fecha" => "2018"
                        "volumen" => "178"
                        "paginaInicial" => "1544"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            77 => array:3 [
              "identificador" => "bib0795"
              "etiqueta" => "78"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Do no harm&#58; a roadmap for responsible machine learning for health care"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [ …6]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1038/s41591-019-0548-6"
                      "Revista" => array:7 [
                        "tituloSerie" => "Nat&#46; Med&#46;"
                        "fecha" => "2019"
                        "volumen" => "25"
                        "paginaInicial" => "1337"
                        "paginaFinal" => "1340"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                        "itemHostRev" => array:3 [
                          "pii" => "S154235651501304X"
                          "estado" => "S300"
                          "issn" => "15423565"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            78 => array:3 [
              "identificador" => "bib0800"
              "etiqueta" => "79"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Clinical management of sepsis can be improved by artificial intelligence&#58; no"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1007/s00134-020-05947-1"
                      "Revista" => array:6 [
                        "tituloSerie" => "Intensive Care Med&#46;"
                        "fecha" => "2020"
                        "volumen" => "3"
                        "paginaInicial" => "1"
                        "paginaFinal" => "3"
                        "itemHostRev" => array:3 [
                          "pii" => "S2468125321000571"
                          "estado" => "S300"
                          "issn" => "24681253"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            79 => array:3 [
              "identificador" => "bib0805"
              "etiqueta" => "80"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Deep medicine&#58; how artificial intelligence can make healthcare human again"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:1 [ …1]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:1 [
                      "Libro" => array:3 [
                        "fecha" => "2019"
                        "editorial" => "Basic Books&#44; Inc"
                        "editorialLocalizacion" => "New York&#44; NY&#44; USA"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            80 => array:3 [
              "identificador" => "bib0810"
              "etiqueta" => "81"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Clinical management of sepsis can be improved by artificial intelligence&#58; yes"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:1 [ …1]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1007/s00134-019-05898-2"
                      "Revista" => array:6 [
                        "tituloSerie" => "Intensive Care Med&#46;"
                        "fecha" => "2020"
                        "volumen" => "46"
                        "paginaInicial" => "375"
                        "paginaFinal" => "377"
                        "link" => array:1 [
                          0 => array:2 [ …2]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
          ]
        ]
      ]
    ]
    "agradecimientos" => array:1 [
      0 => array:4 [
        "identificador" => "xack588981"
        "titulo" => "Acknowledgements"
        "texto" => "<p id="par0315" class="elsevierStylePara elsevierViewall">SING group thanks CITI &#40;<span class="elsevierStyleItalic">Centro de Investigaci&#243;n&#44; Transferencia e Innovaci&#243;n</span>&#41; from University of Vigo for hosting its IT infrastructure&#46;</p>"
        "vista" => "all"
      ]
    ]
  ]
  "idiomaDefecto" => "en"
  "url" => "/02105691/0000004600000003/v1_202202250826/S0210569120301029/v1_202202250826/en/main.assets"
  "Apartado" => array:4 [
    "identificador" => "34"
    "tipo" => "SECCION"
    "es" => array:2 [
      "titulo" => "Revisi&#243;n"
      "idiomaDefecto" => true
    ]
    "idiomaDefecto" => "es"
  ]
  "PDF" => "https://static.elsevier.es/multimedia/02105691/0000004600000003/v1_202202250826/S0210569120301029/v1_202202250826/en/main.pdf?idApp=WMIE&text.app=https://medintensiva.org/"
  "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S0210569120301029?idApp=WMIE"
]
Compartir