Related Experiment Video
Updated: Dec 6, 2025
07:31
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
7.4K
Machine learning based early warning system enables accurate mortality risk prediction for COVID-19
Yue Gao1,2, Guang-Yao Cai1,2, Wei Fang3
1National Medical Center for Major Public Health Events, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, 430000, China.
Nature Communications
|October 7, 2020
Summary
A new mortality risk prediction model for COVID-19 (MRPMC) uses patient data to forecast patient outcomes up to 20 days in advance. This machine learning model helps identify high-risk patients for better health system response.
Area of Science:
- Medical Informatics
- Machine Learning in Healthcare
- Epidemiology
Background:
- Rising COVID-19 cases strain global health systems.
- Increasing COVID-19 mortality necessitates improved patient management.
- Accurate prediction of patient deterioration is crucial for timely intervention.
Purpose of the Study:
- To develop and validate a machine learning model for predicting COVID-19 mortality risk.
- To enable early identification of patients at high risk of physiological deterioration and death.
- To support responsive healthcare strategies for COVID-19 patients.
Main Methods:
- An ensemble machine learning model (MRPMC) was developed using Logistic Regression, Support Vector Machine, Gradient Boosted Decision Tree, and Neural Network.
- The model utilizes patient clinical data available upon admission.
- Validation was performed using internal and two external patient cohorts.
Main Results:
- The MRPMC achieved high predictive performance with AUC values of 0.9621, 0.9760, and 0.9246 in the validation cohorts.
- The model accurately stratifies patients by mortality risk.
- Early prediction of mortality risk is possible up to 20 days in advance.
Conclusions:
- The MRPMC provides a robust tool for expeditious and accurate mortality risk stratification in COVID-19 patients.
- This model can enhance healthcare system responsiveness for high-risk individuals.
- Early risk stratification can lead to improved patient outcomes and resource allocation.
