Medicina Intensiva最新文献_第2页

Application of a machine learning model for early prediction of in-hospital cardiac arrests: Retrospective observational cohort study

IF 2.7 4区医学

Medicina Intensiva Pub Date : 2025-02-01 DOI: 10.1016/j.medin.2024.06.014

L. Socias Crespí , L. Gutiérrez Madroñal , M. Fiorella Sarubbo , M. Borges-Sa , A. Serrano García , D. López Ramos , C. Pruenza Garcia-Hinojosa , E. Martin Garijo

{"title":"Application of a machine learning model for early prediction of in-hospital cardiac arrests: Retrospective observational cohort study","authors":"L. Socias Crespí , L. Gutiérrez Madroñal , M. Fiorella Sarubbo , M. Borges-Sa , A. Serrano García , D. López Ramos , C. Pruenza Garcia-Hinojosa , E. Martin Garijo","doi":"10.1016/j.medin.2024.06.014","DOIUrl":"10.1016/j.medin.2024.06.014","url":null,"abstract":"<div><h3>Objective</h3><div>To describe the results of the application of a Machine Learning (ML) model to predict in-hospital cardiac arrests (ICA) 24 hours in advance in the hospital wards.</div></div><div><h3>Design</h3><div>Retrospective observational cohort study.</div></div><div><h3>Setting</h3><div>Hospital Wards.</div></div><div><h3>Patients</h3><div>Data were extracted from the hospital's Electronic Health Record (EHR). The resulting database contained a total of 750 records corresponding to 620 different patients (370 patients with ICA and 250 control), between may 2009 and december 2021.</div></div><div><h3>Interventions</h3><div>No.</div></div><div><h3>Main variables of interest</h3><div>As predictors of ICA, a set of 28 variables including personal history, vital signs and laboratory data was employed.</div></div><div><h3>Models</h3><div>For the early prediction of ICA, predictive models based on the following ML algorithms and using the mentioned variables, were developed and compared: K Nearest Neighbours, Support Vector Machine, Multilayer Perceptron, Random Forest, Gradient Boosting and Custom Ensemble of Gradient Boosting estimators (CEGB).</div></div><div><h3>Experiments</h3><div>Model training and evaluation was carried out using cross validation. Among metrics of performance, accuracy, specificity, sensitivity and AUC were estimated.</div></div><div><h3>Results</h3><div>The best performance was provided by the CEGB model, which obtained an AUC = 0.90, a specificity = 0.84 and a sensitivity = 0.81. The main variables with influence to predict ICA were level of consciousness, haemoglobin, glucose, urea, blood pressure, heart rate, creatinine, age and hypertension, among others.</div></div><div><h3>Conclusions</h3><div>The use of ML models could be of great support in the early detection of ICA, as the case of the CEGB model endorsed, which enabled good predictions of ICA.</div></div>","PeriodicalId":49268,"journal":{"name":"Medicina Intensiva","volume":"49 2","pages":"Pages 88-95"},"PeriodicalIF":2.7,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143149418","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}