Using Machine Learning Models to Predict Heart Disease and Attack with Non-Clinical and Non-Wearable Variables

This study uses data from the CDC’s Behavioral Risk Factor Surveillance System (BRFSS) to train machine learning models that predict heart disease risk based on non-clinical factors. The research demonstrates the potential for cost-effective, data-driven screening tools for early heart disease detection, though further improvements are needed for real-world application.

Heart disease is the leading cause of death in the United States for over 100 years. However, electrocardiograms (EKG), the primary method of testing for heart issues, costs the average American ~100-5000 dollars per scan. Therefore, there should be other cheaper and time efficient methods for other methods of early onset heart disease detection. We used data from the Behavioral Risk Factor Surveillance System (BRFSS), a telephoned based survey with over 200 thousand participants and partnered with theCenters for Disease Control and Prevention (CDC), to find correlations between heart disease and attacks with non-clinical factors. Multiple machine learning models were trained on BRFSS 2015 data. The best model achieved 70% accuracy, while having a 21% precision and a 79% recall. These results show that while there is potential using non-clinical data combined with machine learning to provide a cost-effective preliminary screening tool for heart disease risk, significant improvements are needed before they can be applied effectively in real-world healthcare.

Explore More!

Research Paper