array:24 [
  "pii" => "S2173572723000504"
  "issn" => "21735727"
  "doi" => "10.1016/j.medine.2023.04.004"
  "estado" => "S300"
  "fechaPublicacion" => "2023-07-01"
  "aid" => "1877"
  "copyright" => "Elsevier España, S.L.U. and SEMICYUC"
  "copyrightAnyo" => "2023"
  "documento" => "simple-article"
  "crossmark" => 1
  "subdocumento" => "cor"
  "cita" => "Med Intensiva. 2023;47:416-7"
  "abierto" => array:3 [
    "ES" => true
    "ES2" => true
    "LATM" => true
  ]
  "gratuito" => true
  "lecturas" => array:1 [
    "total" => 0
  ]
  "Traduccion" => array:1 [
    "en" => array:19 [
      "pii" => "S0210569123000918"
      "issn" => "02105691"
      "doi" => "10.1016/j.medin.2023.04.008"
      "estado" => "S300"
      "fechaPublicacion" => "2023-07-01"
      "aid" => "1877"
      "copyright" => "Elsevier España, S.L.U. y SEMICYUC"
      "documento" => "simple-article"
      "crossmark" => 1
      "subdocumento" => "cor"
      "cita" => "Med Intensiva. 2023;47:416-7"
      "abierto" => array:3 [
        "ES" => true
        "ES2" => true
        "LATM" => true
      ]
      "gratuito" => true
      "lecturas" => array:1 [
        "total" => 0
      ]
      "en" => array:10 [
        "idiomaDefecto" => true
        "cabecera" => "<span class="elsevierStyleTextfn">Letter to the Editor</span>"
        "titulo" => "Machine-learning models for prediction of sepsis patients mortality&#58; A needed consideration"
        "tienePdf" => "en"
        "tieneTextoCompleto" => "en"
        "paginas" => array:1 [
          0 => array:2 [
            "paginaInicial" => "416"
            "paginaFinal" => "417"
          ]
        ]
        "titulosAlternativos" => array:1 [
          "es" => array:1 [
            "titulo" => "Predicci&#243;n de mortalidad en pacientes s&#233;pticos mediante modeles de machine-learning&#58; una reflexi&#243;n necesaria"
          ]
        ]
        "contieneTextoCompleto" => array:1 [
          "en" => true
        ]
        "contienePdf" => array:1 [
          "en" => true
        ]
        "autores" => array:1 [
          0 => array:2 [
            "autoresLista" => "Marcos Valiente Fern&#225;ndez, Francisco de Paula Delgado Moya, Amanda Lesmes Gonz&#225;lez de Aledo, Isa&#237;as Mart&#237;n Bad&#237;a"
            "autores" => array:4 [
              0 => array:2 [
                "nombre" => "Marcos"
                "apellidos" => "Valiente Fern&#225;ndez"
              ]
              1 => array:2 [
                "nombre" => "Francisco de Paula"
                "apellidos" => "Delgado Moya"
              ]
              2 => array:2 [
                "nombre" => "Amanda"
                "apellidos" => "Lesmes Gonz&#225;lez de Aledo"
              ]
              3 => array:2 [
                "nombre" => "Isa&#237;as"
                "apellidos" => "Mart&#237;n Bad&#237;a"
              ]
            ]
          ]
        ]
      ]
      "idiomaDefecto" => "en"
      "Traduccion" => array:1 [
        "en" => array:9 [
          "pii" => "S2173572723000504"
          "doi" => "10.1016/j.medine.2023.04.004"
          "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/S2173572723000504?idApp=WMIE"
        ]
      ]
      "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S0210569123000918?idApp=WMIE"
      "url" => "/02105691/0000004700000007/v1_202306271124/S0210569123000918/v1_202306271124/en/main.assets"
    ]
  ]
  "itemSiguiente" => array:19 [
    "pii" => "S2173572722003022"
    "issn" => "21735727"
    "doi" => "10.1016/j.medine.2022.10.016"
    "estado" => "S300"
    "fechaPublicacion" => "2023-07-01"
    "aid" => "1780"
    "copyright" => "Elsevier Espa&#241;a&#44; S&#46;L&#46;U&#46; and SEMICYUC"
    "documento" => "article"
    "crossmark" => 1
    "subdocumento" => "sco"
    "cita" => "Med Intensiva. 2023;47:418"
    "abierto" => array:3 [
      "ES" => true
      "ES2" => true
      "LATM" => true
    ]
    "gratuito" => true
    "lecturas" => array:1 [
      "total" => 0
    ]
    "en" => array:11 [
      "idiomaDefecto" => true
      "cabecera" => "<span class="elsevierStyleTextfn">Images in Intensive Medicine</span>"
      "titulo" => "Bronchial compression due to dilatated pulmonary artery without pulmonary hypertension"
      "tienePdf" => "en"
      "tieneTextoCompleto" => "en"
      "paginas" => array:1 [
        0 => array:1 [
          "paginaInicial" => "418"
        ]
      ]
      "titulosAlternativos" => array:1 [
        "es" => array:1 [
          "titulo" => "Compresi&#243;n bronquial por arteria pulmonar dilatada sin hipertensi&#243;n pulmonar"
        ]
      ]
      "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" => 753
              "Ancho" => 1013
              "Tamanyo" => 115705
            ]
          ]
          "detalles" => array:1 [
            0 => array:3 [
              "identificador" => "at0010"
              "detalle" => "Figure "
              "rol" => "short"
            ]
          ]
        ]
      ]
      "autores" => array:1 [
        0 => array:2 [
          "autoresLista" => "D&#46;F&#46; L&#243;pez Hormiga, D&#46;M&#46; Combarros M&#233;ndez, I&#46; Keituqwa Y&#225;&#241;ez"
          "autores" => array:3 [
            0 => array:2 [
              "nombre" => "D&#46;F&#46;"
              "apellidos" => "L&#243;pez Hormiga"
            ]
            1 => array:2 [
              "nombre" => "D&#46;M&#46;"
              "apellidos" => "Combarros M&#233;ndez"
            ]
            2 => array:2 [
              "nombre" => "I&#46;"
              "apellidos" => "Keituqwa Y&#225;&#241;ez"
            ]
          ]
        ]
      ]
    ]
    "idiomaDefecto" => "en"
    "Traduccion" => array:1 [
      "es" => array:9 [
        "pii" => "S0210569122000699"
        "doi" => "10.1016/j.medin.2022.03.011"
        "estado" => "S300"
        "subdocumento" => ""
        "abierto" => array:3 [
          "ES" => true
          "ES2" => true
          "LATM" => true
        ]
        "gratuito" => true
        "lecturas" => array:1 [
          "total" => 0
        ]
        "idiomaDefecto" => "es"
        "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S0210569122000699?idApp=WMIE"
      ]
    ]
    "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S2173572722003022?idApp=WMIE"
    "url" => "/21735727/0000004700000007/v1_202306271042/S2173572722003022/v1_202306271042/en/main.assets"
  ]
  "itemAnterior" => array:19 [
    "pii" => "S2173572723000498"
    "issn" => "21735727"
    "doi" => "10.1016/j.medine.2023.04.003"
    "estado" => "S300"
    "fechaPublicacion" => "2023-07-01"
    "aid" => "1868"
    "copyright" => "Elsevier Espa&#241;a&#44; S&#46;L&#46;U&#46; and SEMICYUC"
    "documento" => "simple-article"
    "crossmark" => 1
    "subdocumento" => "crp"
    "cita" => "Med Intensiva. 2023;47:413-5"
    "abierto" => array:3 [
      "ES" => true
      "ES2" => true
      "LATM" => true
    ]
    "gratuito" => true
    "lecturas" => array:1 [
      "total" => 0
    ]
    "en" => array:10 [
      "idiomaDefecto" => true
      "cabecera" => "<span class="elsevierStyleTextfn">Scientific Letter</span>"
      "titulo" => "Validation of the P&#47;FPE index in a cohort of patients with ARDS due to SARS-CoV-2"
      "tienePdf" => "en"
      "tieneTextoCompleto" => "en"
      "paginas" => array:1 [
        0 => array:2 [
          "paginaInicial" => "413"
          "paginaFinal" => "415"
        ]
      ]
      "titulosAlternativos" => array:1 [
        "es" => array:1 [
          "titulo" => "Validaci&#243;n del &#237;ndice P&#47;FPe en una cohorte de enfermos con SDRA secundario a SARS-CoV-2"
        ]
      ]
      "contieneTextoCompleto" => array:1 [
        "en" => true
      ]
      "contienePdf" => array:1 [
        "en" => true
      ]
      "autores" => array:1 [
        0 => array:2 [
          "autoresLista" => "Alejandro Gonz&#225;lez-Castro, Frank Daniel Martos Ben&#237;tez, Alba Fernandez-Rodriguez, Versis Orama-Requejo, Raquel Ferrero-Franco, Yhivian Pe&#241;asco"
          "autores" => array:6 [
            0 => array:2 [
              "nombre" => "Alejandro"
              "apellidos" => "Gonz&#225;lez-Castro"
            ]
            1 => array:2 [
              "nombre" => "Frank Daniel"
              "apellidos" => "Martos Ben&#237;tez"
            ]
            2 => array:2 [
              "nombre" => "Alba"
              "apellidos" => "Fernandez-Rodriguez"
            ]
            3 => array:2 [
              "nombre" => "Versis"
              "apellidos" => "Orama-Requejo"
            ]
            4 => array:2 [
              "nombre" => "Raquel"
              "apellidos" => "Ferrero-Franco"
            ]
            5 => array:2 [
              "nombre" => "Yhivian"
              "apellidos" => "Pe&#241;asco"
            ]
          ]
        ]
      ]
    ]
    "idiomaDefecto" => "en"
    "Traduccion" => array:1 [
      "es" => array:9 [
        "pii" => "S0210569123000839"
        "doi" => "10.1016/j.medin.2023.03.008"
        "estado" => "S300"
        "subdocumento" => ""
        "abierto" => array:3 [
          "ES" => true
          "ES2" => true
          "LATM" => true
        ]
        "gratuito" => true
        "lecturas" => array:1 [
          "total" => 0
        ]
        "idiomaDefecto" => "es"
        "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S0210569123000839?idApp=WMIE"
      ]
    ]
    "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S2173572723000498?idApp=WMIE"
    "url" => "/21735727/0000004700000007/v1_202306271042/S2173572723000498/v1_202306271042/en/main.assets"
  ]
  "en" => array:13 [
    "idiomaDefecto" => true
    "cabecera" => "<span class="elsevierStyleTextfn">Letter to the Editor</span>"
    "titulo" => "Machine-learning models for prediction of sepsis patients mortality&#58; A needed consideration"
    "tieneTextoCompleto" => true
    "saludo" => "Dear Editor&#58;"
    "paginas" => array:1 [
      0 => array:2 [
        "paginaInicial" => "416"
        "paginaFinal" => "417"
      ]
    ]
    "autores" => array:1 [
      0 => array:4 [
        "autoresLista" => "Marcos Valiente Fern&#225;ndez, Francisco de Paula Delgado Moya, Amanda Lesmes Gonz&#225;lez de Aledo, Isa&#237;as Mart&#237;n Bad&#237;a"
        "autores" => array:4 [
          0 => array:4 [
            "nombre" => "Marcos"
            "apellidos" => "Valiente Fern&#225;ndez"
            "email" => array:1 [
              0 => "mvalientefernandez@gmail.com"
            ]
            "referencia" => array:1 [
              0 => array:2 [
                "etiqueta" => "<span class="elsevierStyleSup">&#42;</span>"
                "identificador" => "cor0005"
              ]
            ]
          ]
          1 => array:2 [
            "nombre" => "Francisco de Paula"
            "apellidos" => "Delgado Moya"
          ]
          2 => array:2 [
            "nombre" => "Amanda"
            "apellidos" => "Lesmes Gonz&#225;lez de Aledo"
          ]
          3 => array:2 [
            "nombre" => "Isa&#237;as"
            "apellidos" => "Mart&#237;n Bad&#237;a"
          ]
        ]
        "afiliaciones" => array:1 [
          0 => array:2 [
            "entidad" => "Servicio de Medicina Intensiva&#44; Hospital Universitario 12 de Octubre&#44; Avenida de C&#243;rdoba&#44; s&#47;n&#44; 28041&#44; Madrid&#44; Spain"
            "identificador" => "aff0005"
          ]
        ]
        "correspondencia" => array:1 [
          0 => array:3 [
            "identificador" => "cor0005"
            "etiqueta" => "&#8270;"
            "correspondencia" => "Corresponding author&#46;"
          ]
        ]
      ]
    ]
    "titulosAlternativos" => array:1 [
      "es" => array:1 [
        "titulo" => "Predicci&#243;n de mortalidad en pacientes s&#233;pticos mediante modeles de machine-learning&#58; una reflexi&#243;n necesaria"
      ]
    ]
    "textoCompleto" => "<span class="elsevierStyleSections"><p id="par0005" class="elsevierStylePara elsevierViewall">The article &#8220;Machine-learning models for prediction of sepsis patients mortality&#8221;<a class="elsevierStyleCrossRef" href="#bib0005"><span class="elsevierStyleSup">1</span></a> is very interesting for many reasons&#58; the methodology used&#44; the adoption of advanced stadisticals technics&#44; the huge sample size&#44; and the aim of universalize the results comparing different populations&#46; However&#44; for us&#44; it is remarkable that they do not emphasize the struggle between the clinical prediction and the model prediction&#46;</p><p id="par0010" class="elsevierStylePara elsevierViewall">QuickSOFA &#40;QS&#41; scale intend to predict mortality as a brief version of a more complex scale&#44; thresh from the Sepsis-3 recommendations&#46; It is based on 3 variables&#58; respiratory rate &#40;RR&#41;&#44; neurological impairment &#40;following the Glasgow coma scale&#58; GCS&#41;&#44; and systolic blood pressure &#40;SBP&#41;<span class="elsevierStyleHsp" style=""></span>&#60;<span class="elsevierStyleHsp" style=""></span>100<span class="elsevierStyleHsp" style=""></span>mmHg&#46;</p><p id="par0015" class="elsevierStylePara elsevierViewall">As it is said in the paper&#44; the SOFA scale had not had the good clinical prediction that it has in the model&#46; In the search of why this could happen&#44; if we focused on the variables included&#44; they were some of the least important in the predictive model&#58; RR &#40;13th out of 15&#41;&#44; GCS &#40;9th out of 15&#41; and SBP &#40;15th out of 15&#41;&#46; The most exciting point is that there are other accessible and more valuable variables for the model at our reach&#44; as age&#44; median blood pressure&#44; and temperature&#46;</p><p id="par0020" class="elsevierStylePara elsevierViewall">The interesting thing is that there are easily accessible variables that are more relevant to the model&#44; such as age&#44; mean blood pressure&#44; and temperature&#44; than those used by QS&#46;</p><p id="par0025" class="elsevierStylePara elsevierViewall">The importance of variables is not entirely equivalent to the punctuation in the logistic regression results&#59; it is more related to the need for that variable for decision-making by the algorithm&#46; Therefore&#44; a frequently used variable is relevant when predicting and a less frequently used variable does not contribute as much&#46;</p><p id="par0030" class="elsevierStylePara elsevierViewall">Here lies the crucial issue&#46; Traditionally&#44; we have made use of predictors based on linear statistical models &#40;multivariate logistic regression&#41; that have offered us predictive variables which allow us to detect and grade the risk of patients in an objective manner&#46; Currently&#44; machine learning techniques - based on non-linear statistical models &#8211; grant us a greater predictive capacity with a different priority of variables than we knew&#46;<a class="elsevierStyleCrossRef" href="#bib0010"><span class="elsevierStyleSup">2</span></a></p><p id="par0035" class="elsevierStylePara elsevierViewall">This represents a key change&#44; as it not only implies the generation of &#8220;new and different prediction for the same patient&#44; but also forces us to think of new predictors for diseases that we already thought we understood enough&#46;<a class="elsevierStyleCrossRef" href="#bib0015"><span class="elsevierStyleSup">3</span></a> The predictive success of these models should encourage us to create datasets &#40;collaborative databases&#41; that allow us to exploit their predictive and hierarchical potential of the variables employed&#44; which would lead us to prioritize areas of knowledge in diseases that we may have been neglecting&#46;</p><p id="par0040" class="elsevierStylePara elsevierViewall">We don&#39;t just have traditional statistics&#44; but we have a multi-purpose tool machine learning&#46; Our responsibility is to know how to use all the tools we have properly&#44; to identify the problem we want to solve and therefore choose the appropriate statistical tool&#46;<a class="elsevierStyleCrossRef" href="#bib0020"><span class="elsevierStyleSup">4</span></a> This is not an invitation to give up on tradition&#44; but rather a need to include new statistical tools to improve our knowledge in a complex clinical scenario&#46;<a class="elsevierStyleCrossRef" href="#bib0025"><span class="elsevierStyleSup">5</span></a></p><span id="sec0005" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0005">Financing</span><p id="par0045" class="elsevierStylePara elsevierViewall">None&#46;</p></span><span id="sec0010" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0010">Conflicts of interest</span><p id="par0050" class="elsevierStylePara elsevierViewall">The authors declare that none have conflicts of interest&#46;</p></span></span>"
    "textoCompletoSecciones" => array:1 [
      "secciones" => array:3 [
        0 => array:2 [
          "identificador" => "sec0005"
          "titulo" => "Financing"
        ]
        1 => array:2 [
          "identificador" => "sec0010"
          "titulo" => "Conflicts of interest"
        ]
        2 => array:1 [
          "titulo" => "References"
        ]
      ]
    ]
    "pdfFichero" => "main.pdf"
    "tienePdf" => true
    "bibliografia" => array:2 [
      "titulo" => "References"
      "seccion" => array:1 [
        0 => array:2 [
          "identificador" => "bibs0005"
          "bibliografiaReferencia" => array:5 [
            0 => array:3 [
              "identificador" => "bib0005"
              "etiqueta" => "1"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Machine-learning models for prediction of sepsis patients mortality"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:3 [
                            0 => "C&#46; Bao"
                            1 => "F&#46; Deng"
                            2 => "S&#46; Zhao"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1016/j.medine.2022.06.024"
                      "Revista" => array:3 [
                        "tituloSerie" => "Med Intensiva &#40;Engl Ed&#41;&#46;"
                        "fecha" => "2022"
                        "numero" => "Nov 4"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            1 => array:3 [
              "identificador" => "bib0010"
              "etiqueta" => "2"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "The unreasonable effectiveness of data"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:3 [
                            0 => "A&#46; Halevy"
                            1 => "P&#46; Norvig"
                            2 => "F&#46; Pereira"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:1 [
                      "Revista" => array:6 [
                        "tituloSerie" => "IEEE Intell Syst&#46;"
                        "fecha" => "2009"
                        "volumen" => "24"
                        "numero" => "2"
                        "paginaInicial" => "8"
                        "paginaFinal" => "12"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            2 => array:3 [
              "identificador" => "bib0015"
              "etiqueta" => "3"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Statistical modeling&#58; the two cultures &#40;with discussion&#41;"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => false
                          "autores" => array:1 [
                            0 => "L&#46; Breiman"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:1 [
                      "Revista" => array:5 [
                        "tituloSerie" => "Stat Sci&#46;"
                        "fecha" => "2001"
                        "volumen" => "16"
                        "paginaInicial" => "199"
                        "paginaFinal" => "231"
                      ]
                    ]
                  ]
                ]
              ]
            ]
            3 => array:3 [
              "identificador" => "bib0020"
              "etiqueta" => "4"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "A comparison of machine learning models versus clinical evaluation for mortality prediction in patients with sepsis"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:5 [
                            0 => "W&#46;P&#46;T&#46;M&#46; van Doorn"
                            1 => "P&#46;M&#46; Stassen"
                            2 => "H&#46;F&#46; Borggreve"
                            3 => "M&#46;J&#46; Schalkwijk"
                            4 => "J&#46; Stoffers"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1371/journal.pone.0245157"
                      "Revista" => array:5 [
                        "tituloSerie" => "PLoS One&#46;"
                        "fecha" => "2021"
                        "volumen" => "16"
                        "numero" => "1"
                        "itemHostRev" => array:3 [
                          "pii" => "S0091674910005907"
                          "estado" => "S300"
                          "issn" => "00916749"
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
            4 => array:3 [
              "identificador" => "bib0025"
              "etiqueta" => "5"
              "referencia" => array:1 [
                0 => array:2 [
                  "contribucion" => array:1 [
                    0 => array:2 [
                      "titulo" => "Assessing clinical heterogeneity in sepsis through treatment patterns and machine learning"
                      "autores" => array:1 [
                        0 => array:2 [
                          "etal" => true
                          "autores" => array:6 [
                            0 => "A&#46;E&#46; Fohner"
                            1 => "J&#46;D&#46; Greene"
                            2 => "B&#46;L&#46; Lawson"
                            3 => "J&#46;H&#46; Chen"
                            4 => "P&#46; Kipnis"
                            5 => "G&#46;J&#46; Escobar"
                          ]
                        ]
                      ]
                    ]
                  ]
                  "host" => array:1 [
                    0 => array:2 [
                      "doi" => "10.1093/jamia/ocz106"
                      "Revista" => array:7 [
                        "tituloSerie" => "J Am Med Inform Assoc&#46;"
                        "fecha" => "2019"
                        "volumen" => "26"
                        "numero" => "Dec &#40;12&#41;"
                        "paginaInicial" => "1466"
                        "paginaFinal" => "1477"
                        "link" => array:1 [
                          0 => array:2 [
                            "url" => "https://www.ncbi.nlm.nih.gov/pubmed/31314892"
                            "web" => "Medline"
                          ]
                        ]
                      ]
                    ]
                  ]
                ]
              ]
            ]
          ]
        ]
      ]
    ]
  ]
  "idiomaDefecto" => "en"
  "url" => "/21735727/0000004700000007/v1_202306271042/S2173572723000504/v1_202306271042/en/main.assets"
  "Apartado" => array:4 [
    "identificador" => "64604"
    "tipo" => "SECCION"
    "en" => array:2 [
      "titulo" => "Letters to the Editor"
      "idiomaDefecto" => true
    ]
    "idiomaDefecto" => "en"
  ]
  "PDF" => "https://static.elsevier.es/multimedia/21735727/0000004700000007/v1_202306271042/S2173572723000504/v1_202306271042/en/main.pdf?idApp=WMIE&text.app=https://medintensiva.org/"
  "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S2173572723000504?idApp=WMIE"
]
Share
Journal Information
Vol. 47. Issue 7.
Pages 416-417 (July 2023)
Share
Share
Download PDF
More article options
Vol. 47. Issue 7.
Pages 416-417 (July 2023)
Letter to the Editor
Full text access
Machine-learning models for prediction of sepsis patients mortality: A needed consideration
Predicción de mortalidad en pacientes sépticos mediante modeles de machine-learning: una reflexión necesaria
Visits
1066
Marcos Valiente Fernández
Corresponding author
mvalientefernandez@gmail.com

Corresponding author.
, Francisco de Paula Delgado Moya, Amanda Lesmes González de Aledo, Isaías Martín Badía
Servicio de Medicina Intensiva, Hospital Universitario 12 de Octubre, Avenida de Córdoba, s/n, 28041, Madrid, Spain
This item has received
Article information
Full Text
Bibliography
Download PDF
Statistics
Full Text
Dear Editor:

The article “Machine-learning models for prediction of sepsis patients mortality”1 is very interesting for many reasons: the methodology used, the adoption of advanced stadisticals technics, the huge sample size, and the aim of universalize the results comparing different populations. However, for us, it is remarkable that they do not emphasize the struggle between the clinical prediction and the model prediction.

QuickSOFA (QS) scale intend to predict mortality as a brief version of a more complex scale, thresh from the Sepsis-3 recommendations. It is based on 3 variables: respiratory rate (RR), neurological impairment (following the Glasgow coma scale: GCS), and systolic blood pressure (SBP)<100mmHg.

As it is said in the paper, the SOFA scale had not had the good clinical prediction that it has in the model. In the search of why this could happen, if we focused on the variables included, they were some of the least important in the predictive model: RR (13th out of 15), GCS (9th out of 15) and SBP (15th out of 15). The most exciting point is that there are other accessible and more valuable variables for the model at our reach, as age, median blood pressure, and temperature.

The interesting thing is that there are easily accessible variables that are more relevant to the model, such as age, mean blood pressure, and temperature, than those used by QS.

The importance of variables is not entirely equivalent to the punctuation in the logistic regression results; it is more related to the need for that variable for decision-making by the algorithm. Therefore, a frequently used variable is relevant when predicting and a less frequently used variable does not contribute as much.

Here lies the crucial issue. Traditionally, we have made use of predictors based on linear statistical models (multivariate logistic regression) that have offered us predictive variables which allow us to detect and grade the risk of patients in an objective manner. Currently, machine learning techniques - based on non-linear statistical models – grant us a greater predictive capacity with a different priority of variables than we knew.2

This represents a key change, as it not only implies the generation of “new and different prediction for the same patient, but also forces us to think of new predictors for diseases that we already thought we understood enough.3 The predictive success of these models should encourage us to create datasets (collaborative databases) that allow us to exploit their predictive and hierarchical potential of the variables employed, which would lead us to prioritize areas of knowledge in diseases that we may have been neglecting.

We don't just have traditional statistics, but we have a multi-purpose tool machine learning. Our responsibility is to know how to use all the tools we have properly, to identify the problem we want to solve and therefore choose the appropriate statistical tool.4 This is not an invitation to give up on tradition, but rather a need to include new statistical tools to improve our knowledge in a complex clinical scenario.5

Financing

None.

Conflicts of interest

The authors declare that none have conflicts of interest.

References
[1]
C. Bao, F. Deng, S. Zhao.
Machine-learning models for prediction of sepsis patients mortality.
Med Intensiva (Engl Ed)., (2022),
[2]
A. Halevy, P. Norvig, F. Pereira.
The unreasonable effectiveness of data.
IEEE Intell Syst., 24 (2009), pp. 8-12
[3]
L. Breiman.
Statistical modeling: the two cultures (with discussion).
Stat Sci., 16 (2001), pp. 199-231
[4]
W.P.T.M. van Doorn, P.M. Stassen, H.F. Borggreve, M.J. Schalkwijk, J. Stoffers, et al.
A comparison of machine learning models versus clinical evaluation for mortality prediction in patients with sepsis.
[5]
A.E. Fohner, J.D. Greene, B.L. Lawson, J.H. Chen, P. Kipnis, G.J. Escobar, et al.
Assessing clinical heterogeneity in sepsis through treatment patterns and machine learning.
J Am Med Inform Assoc., 26 (2019), pp. 1466-1477
Copyright © 2023. Elsevier España, S.L.U. and SEMICYUC
Download PDF
Idiomas
Medicina Intensiva (English Edition)
Article options
Tools
es en

¿Es usted profesional sanitario apto para prescribir o dispensar medicamentos?

Are you a health professional able to prescribe or dispense drugs?