Risk assessment in COVID-19 patients: A multiclass classification approach. (2022)

• Record Type: Journal Article
• Authors: Bárcenas, Roberto; Fuentes-García, Ruth
• Is Part Of: Informatics in medicine unlocked. Volume 32(2022)
• Publication Date: 2022
• Subjects: COVID-19 -- Machine Learning -- Risk assessment -- Multiclass classification -- Feature importance; Medical informatics -- Periodicals; 610.285
• DOI: 10.1016/j.imu.2022.101023
• Languages: English
• ISSNs: 2352-9148

Abstract

Abstract: Understanding SARS-CoV-2 infection that causes COVID-19 disease among the population was fundamental to determine the risk factors associated with severe cases or even death. Amidst the study of the pandemic, Artificial Intelligence (AI) and Machine Learning (ML) have been successfully applied in many areas such as biomedicine. Using a dataset from the Mexican Ministry of Health, we performed a multiclass classification scheme for the detection of risks in COVID-19 patients and implemented three Machine Learning algorithms achieving the following accuracy measures: Random Forest (89.86%), GBM (89.37%) XGBoost (89.97%). The key findings are the identification of relevant components associated with different severities of COVID-19 disease. Among these factors, we found sex, age, days elapsed from the beginning of symptoms, symptoms such as dyspnea and polypnea; and other comorbidities such as diabetes and hypertension. This setting allows us to establish predicting algorithms to model the risk that an individual or a specific group of people face after contracting COVID-19 and the factors associated with developing complications or receiving appropriate treatment. Highlights: Machine Learning is key to identifying the causes of risk for severe COVID-19. Bagging and boosting algorithms provide a suitable way of multiclass classification. COVID-19 risk assessment can be implemented for vaccination and lock-down strategies.