eLife | |
Development and evaluation of a machine learning-based in-hospital COVID-19 disease outcome predictor (CODOP): A multicontinental retrospective study | |
Miguel Pedrera-Jiménez1  Noelia García Barrio1  Rosmery Gross Artega2  Juan Ignacio Ramirez3  Pablo Young3  Francisco Rivas-Ruiz4  Bruno Boietti5  Javier A Pollan5  Ivan A Huespe5  Florencia Pugliese6  Pascual Ruben Valdez6  Rosa Castagna6  Benjamin Leiding7  Magdy Teresa Canales Beltrán8  Carlos Lumbreras9  Antonio Lalueza Blanco9  José Manuel Ramos-Rincón1,10  Jesús Millán Núñez-Cortés1,11  María Dolores Martin-Escalante1,12  José Manuel Casas-Rojo1,13  Juan Miguel Antón-Santos1,13  Ricardo Gómez-Huelgas1,14  Estela Edith Titto Omonte1,15  Nico Funke1,16  David Gómez-Varela1,16  Disha Purohit1,17  Maria Ángeles Onieva-García1,18  Riku Klén1,19  | |
[1] Data Science Unit, Research Institute Hospital 12 de Octubre, Madrid, Spain;Epidemiology Unit, Hospital of San Juan de Dios, Santa Cruz, Bolivia;Hospital Británico of Buenos Aires, Buenos Aires, Argentina;Hospital Costa del Sol. Research Unit, Marbella, Spain;Hospital Italiano de Buenos Aires, Buenos Aires, Argentina;Hospital Velez Sarsfield, Buenos Aires, Argentina;Institute for Software and Systems Engineering at TU Clausthal, Clausthal, Germany;Instituto Hondureno of social security, Hospital Honduras Medical Centre, Tegucigalpa, Honduras;Internal Medicine Department, 12 de Octubre University Hospital, Madrid, Spain;Internal Medicine Department, General University Hospital of Alicante, Alicante Institute for 22 Health and Biomedical Research (ISABIAL), Alicante, Spain;Internal Medicine Department, Gregorio Marañón University Hospital, Madrid, Spain;Internal Medicine Department, Hospital Costa del Sol, Marbella, Spain;Internal Medicine Department, Infanta Cristina University Hospital, Madrid, Spain;Internal Medicine Department, Regional University Hospital of Málaga, Biomedical Research Institute of Málaga (IBIMA), University of Málaga (UMA), Málaga, Spain;Internal Medicine Service, Hospital Santa Cruz - Caja Petrolera de Salud, Santa Cruz, Bolivia;Max Planck Institute for Experimental Medicine, Göttingen, Germany;Max Planck Institute of Experimental Medicine, Göttingen, Germany;Preventive Medicine Department, Hospital Costa del Sol, Marbella, Spain;Turku PET Centre, University of Turku and Turku University Hospital, Turku, Finland; | |
关键词: COVID-19; machine-learning; prediction; triage; | |
DOI : 10.7554/eLife.75985 | |
来源: DOAJ |
【 摘 要 】
New SARS-CoV-2 variants, breakthrough infections, waning immunity, and sub-optimal vaccination rates account for surges of hospitalizations and deaths. There is an urgent need for clinically valuable and generalizable triage tools assisting the allocation of hospital resources, particularly in resource-limited countries. We developed and validate CODOP, a machine learning-based tool for predicting the clinical outcome of hospitalized COVID-19 patients. CODOP was trained, tested and validated with six cohorts encompassing 29223 COVID-19 patients from more than 150 hospitals in Spain, the USA and Latin America during 2020–22. CODOP uses 12 clinical parameters commonly measured at hospital admission for reaching high discriminative ability up to 9 days before clinical resolution (AUROC: 0·90–0·96), it is well calibrated, and it enables an effective dynamic risk stratification during hospitalization. Furthermore, CODOP maintains its predictive ability independently of the virus variant and the vaccination status. To reckon with the fluctuating pressure levels in hospitals during the pandemic, we offer two online CODOP calculators, suited for undertriage or overtriage scenarios, validated with a cohort of patients from 42 hospitals in three Latin American countries (78–100% sensitivity and 89–97% specificity). The performance of CODOP in heterogeneous and geographically disperse patient cohorts and the easiness of use strongly suggest its clinical utility, particularly in resource-limited countries.
【 授权许可】
Unknown