A Machine Learning Approach to Determine Risk Factors for Respiratory Bacterial/Fungal Coinfection in Critically Ill Patients - A Spanish Perspective

Aug 6, 2026

Upload by Eliana

Bacterial/fungal coinfections are linked to antibiotic overuse and worse critical-care outcomes, including prolonged ICU stay and higher mortality. The study develops machine learning–based predictive models to recognize respiratory bacterial or fungal coinfection at ICU admission in patients with influenza A and SARS-CoV-2. Secondary analysis covers two prospective multicenter cohorts; multiple logistic regression and random forests identify associated factors, with performance evaluated using ROC AUC and out-of-bag validation.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages17

Reading Time43 min

AccessFree to download

Abstract

Background

Methods

Results

antibiotics Article A Machine Learning Approach to Determine Risk Factors for Respiratory Bacterial/Fungal Coinfection in Critically Ill Patients with Influenza and SARS-CoV-2 Infection: A Spanish Perspective Alejandro Rodríguez 1,2,3,4, *,†, Josep Gómez 2,5,†, Ignacio Martín-Loeches 6, Laura Claverias 1,3, Emili Díaz 7,8, Rafael Zaragoza 9, Marcio Borges-Sa 10, Frederic Gómez-Bertomeu 2,3,11,12, Álvaro Franquet 2,5, Sandra Trefler 1,3, Carlos González Garzón 13, Lissett Cortés 13, Florencia Alés 14, Susana Sancho 15, Jordi Solé-Violán 16, Ángel Estella 17,18, Julen Berrueta 1,19, Alejandro García-Martínez 1,19, Borja Suberviola 20, Juan J. Guardiola 21 and María Bodí 1,2,3 on behalf of GETGAG/COVID-19 SEMICYUC Working Group Citation: Rodríguez, A.; Gómez, J.; Martín-Loeches, I.; Claverias, L.; Díaz, E.; Zaragoza, R.; Borges-Sa, M.; Gómez-Bertomeu, F.; Franquet, Á .; Trefler, S.; et al. A Machine Learning Approach to Determine Risk Factors for Respiratory Bacterial/Fungal Coinfection in Critically Ill Patients with Influenza and SARS-CoV-2 Infection: A Spanish Perspective. Antibiotics 2024, 13, 968. [https://](https://) [doi.org/10.3390/antibiotics13100968](doi.org/10.3390/antibiotics13100968)[ ](doi.org/10.3390/antibiotics13100968)Academic Editor: Giovanna Travi Received: 13 September 2024 Revised: 9 October 2024 Accepted: 10 October 2024 Published: 14 October 2024 Copyright: © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . 1 Critical Care Department, Hospital Universitari Joan XXIII, 43005 Tarragona, Spain; lauraclaverias@gmail.com (L.C.); sitrefler@yahoo.es (S.T.); julen.berrueta@estudiants.urv.cat (J.B.); alejgarcia.hj23.ics@gencat.cat (A.G.-M.); mbodi.hj23.ics@gencat.cat (M.B.) 2 Faculty of Medicine, Universitat Rovira & Virgili, 43005 Tarragona, Spain; [josep.goal@gmail.com](josep.goal@gmail.com) (J.G.); ffgomez.hj23 .ics@gencat.cat (F.G.-B.); afranquet.hj23 .ics@gencat.cat (Á .F.) 3 Pere Virgili Health Research Institute, 43005 Tarragona, Spain 4 Centre for Biomedical Research Network Respiratory Diseases (CIBERES), 43005 Tarragona, Spain 5 Technical Secretary’s Department, Hospital Universitari Joan XXIII, 43005 Tarragona, Spain 6 Department of Intensive Care Medicine, Multidisciplinary Intensive Care Research Organization (MICRO), St James’ Hospita, D08 NHY1 Dublin, Ireland; [drmartinloeches@gmail.com](drmartinloeches@gmail.com) 7 Critical Care Department, Hospital Universitari Parc Tauli, 08208 Sabadell, Spain; [emilio.diaz.santos@gmail.com](emilio.diaz.santos@gmail.com) 8 Medicine Faculty, Universitat Autónoma de Barcelona, 08193 Sabadell, Spain 9 Critical Care Department, Hospital Dr. Peset, 46017 Valencia, Spain; zaragoza_[raf@gva.es](raf@gva.es) 10 Critical Care Department, Hospital Son Llatzer, 07198 Palma de Mallorca, Spain; [marcio.borges.sa1967@gmail.com](marcio.borges.sa1967@gmail.com) 11 Microbiology/Clinical Analysis Laboratory, Hospital Universitari de Tarragona Joan XXIII, 43005 Tarragona, Spain 12 Center for Biomedical Research in Infectious Diseases Network (CIBERINFEC), 28220 Madrid, Spain 13 Postgrado Medicina Crítica y Cuidado Intensivo, Facultad de Medicina, Fundación Universitari Ciencias dela Salud, Distrito Especial, Cra. 54 No.67A-80, Bogotá 111221, Colombia; [carlosmariogonzalezgarzon@hotmail.com](carlosmariogonzalezgarzon@hotmail.com) (C.G.G.); [llcortes@fucsalud.edu.co](llcortes@fucsalud.edu.co) (L.C.) 14 Internal Medicine Department, Hospital Dr. Alejandro Gutiérrez, Venado Tuerto S2600, Argentina; [mafloales@gmail.com](mafloales@gmail.com) 15 Critical Care Department, Hospital Universitrio y Politécnico La Fe, 46026 Valencia, Spain

FAQs

What is the document’s main objective?

To develop machine learning predictive models that identify respiratory bacterial or fungal coinfections upon ICU admission in critically ill patients with influenza and SARS-CoV-2 infection.

Which modeling methods are used?

The study uses multiple logistic regression (MLR) and random forest (RF) to identify factors associated with bacterial/fungal coinfection in the overall population and in influenza and COVID-19 subgroups.

How is model performance assessed?

Performance is evaluated using ROC area under the curve (AUC) and out-of-bag (OOB) validation for the respective models.

Aug 6, 2026

Upload by Eliana