Predicting the outcome for COVID-19 patients by applying time series classification to electronic health records

Abstract

Background COVID-19 caused more than 622 thousand deaths in Brazil. The infection can be asymptomatic and cause mild symptoms, but it also can evolve into a severe disease and lead to death. It is difficult to predict which patients will develop severe disease. There are, in the literature, machine learning models capable of assisting diagnose and predicting outcomes for several diseases, but usually these models require laboratory tests and/or imaging. Methods We conducted a observational cohort study that evaluated vital signs and measurements from patients who were admitted to Hospital das Clínicas (São Paulo, Brazil) between March 2020 and October 2021 due to COVID-19. The data was then represented as univariate and multivariate time series, that were used to train and test machine learning models capable of predicting a patient’s outcome. Results Time series-based machine learning models are capable of predicting a COVID-19 patient’s outcome with up to 96% general accuracy and 81% accuracy considering only the first hospitalization day. The models can reach up to 99% sensitivity (discharge prediction) and up to 91% specificity (death prediction). Conclusions Results indicate that time series-based machine learning models combined with easily obtainable data can predict COVID-19 outcomes and support clinical decisions. With further research, these models can potentially help doctors diagnose other diseases.

Data-driven, cross-disciplinary collaboration: lessons learned at the largest academic health center in Latin America during the COVID-19 pandemic

“…Implementation of EHR brings healthcare closer to data science, computational biology, and artificial intelligence ( 10 ). In our initiatives, we applied artificial intelligence and contemporary computational methods to analyze hospitalization data through collaborations with computer science groups ( 30 , 48 , 50 ). Caution is advised regarding such secondary uses of healthcare data from EHR due to potential misinterpretation and concerns about data quality, especially missing or inaccurate data ( 72 ).…”

An Ensemble Method for Determining the Importance of Selected Social and Health Factors Affecting the Inpatient Treatment Quality

“…1 Literature review, in the literature we could find a lot of work that deals with the application of classification methods in determining the influence of different types of factors, from medical and social to economic, including, in addition to classification, forecasting and prediction [70]. F It could be also used in the diagnosis and prediction of the development of various diseases, such as breast 6 cancer in [71], HIV in [72], COVID19 in [73] etc. but the authors have not found any application of the proposed generic model.…”

Patient allocation method in major epidemics under the situation of hierarchical diagnosis and treatment

“…Studies related to major pandemics cover a wide range, including vaccine development [ 7 – 9 ], epidemic forecasting [ 10 , 11 ], and mitigation strategy [ 12 , 13 ]. The latter is an important research area and includes isolation [ 14 ], protection [ 15 , 16 ], prevention and control [ 17 ], vaccination [ 18 – 20 ].…”