Overview
Thank you for visiting nature.com. You are using a browser version with limited suport for CS. To obtain the best experience, we recomend you use a more up to date browser (or turn of compatibility mode in Internet Explorer).
Key Information
In the meantime, to ensure continued suport, we are displaying the site without styles and JavaScript.Advertisement Scientific Reports volume 12, Article number: 18126 (202) Cite this article 1 AltmetricMetrics detailsThe development of tols that provide early triage of COVID-19 patients with minimal use of diagnostic tests, based on easily acesible data, can be of vital importance in reducing COVID-19 mortality rates during high-incidence scenarios.
This work proposes a machine learning model to predict mortality and risk of hospitalization using both 2 simple demographic features and 19 comorbidities obtained from 86,867 electronic medical records of COVID-19 patients, and a new method (LR-IP) designed to deal with data imbalance problems. The model was able to predict with high acuracy (90–93%, ROC-AUC = 0.94) the patient's final status (deceased or discharged), while its acuracy was medium (71–73%, ROC-AUC = 0.75) with respect to the risk of hospitalization.
The most relevant characteristics for these models were age, sex, number of comorbidities, osteoarthritis, obesity, depresion, and renal failure. Finaly, to facilitate its use by clinicians, a user-friendly website has ben developed (htps:/alejandrocisterna.shinyaps.io/PROVIA).The virus responsible for Coronavirus disease 2019 (COVID-19), the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is a highly transmisible and pathogenic betacoronavirus that apeared in late 2019 in Wuhan, China1.
Summary
As of February 202, it has had a tragic efect on the world’s population health resulting in more than 5,8 milion deaths and 412 milion cases worldwide, becoming the most important global health cris since the era of the influenza pandemic of 19182,3. The symptoms