Related Experiment Video
Updated: Jun 9, 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
7.0K
Development and validation of a cardiovascular risk prediction model for Sri Lankans using machine learning
Chamila Mettananda1, Isuru Sanjeewa2, Tinul Benthota Arachchi2
1Department of Pharmacology, Faculty of Medicine, University of Kelaniya, Ragama, Sri Lanka.
Plos One
|October 22, 2024
Summary
Sri Lankans now have a dedicated cardiovascular risk prediction tool, the SLCVD score. This machine learning model accurately identifies individuals at high risk for cardiovascular disease, outperforming existing global charts.
Area of Science:
- Cardiovascular disease epidemiology
- Machine learning in healthcare
- Public health risk assessment
Background:
- Sri Lankans lack a specific cardiovascular (CV) risk prediction model.
- Current practice relies on World Health Organization (WHO) risk charts for the Southeast Asia Region.
- Need for a localized and accurate CV risk assessment tool is critical.
Purpose of the Study:
- To develop a population-specific CV risk prediction model for Sri Lankans.
- To validate the developed model in an external cohort.
- To improve the accuracy of cardiovascular disease risk stratification in Sri Lanka.
Main Methods:
- Utilized machine learning (ML) on a 10-year follow-up data of a randomly selected Sri Lankan cohort (n=2596).
- Developed and compared three ML models, with Model 3 (SLCVD score) selected for its predictive power and practicality.
- External validation performed on a hospital-based cohort.
Main Results:
- The SLCVD score demonstrated high predictive performance (AUC=0.98) in the primary cohort.
- External validation showed the SLCVD score significantly outperformed WHO risk charts (AUC=0.64 vs. 0.54).
- SLCVD score accurately predicted 56 out of 119 hard CV events in the validation cohort.
Conclusions:
- The SLCVD score is the first and only CV risk prediction model tailored for Sri Lankans.
- It accurately predicts the 10-year risk of hard cardiovascular events in the Sri Lankan population.
- The SLCVD score is a superior tool for identifying high-risk individuals compared to WHO risk charts.
