Explainable Artificial Intelligence (XAI) Approach to Heart Disease Prediction

Abstract

Cardiovascular continue to stand as the leading cause of mortality globally. The World Health Organization (WHO) noted that in 2019, 32% of all fatalities were attributed to heart-related issues. In India, the Ministry of Health and Family Welfare reported that cardiovascular diseases accounted for 28.1% of the overall deaths. [1] Approximately 85% of individuals diagnosed with heart failure survive the first year after diagnosis, with rates decreasing to around 55% at two years, approximately 33% at five years, and about 35% at ten years. [2] Cardiovascular diseases, in 2019, claimed the lives of approximately 17.9 million people worldwide, making them the primary cause of death, constituting around 32% of all global deaths. [3] Underscoring the critical need for predictive models in early detection, our proposal integrates machine learning with Explainable AI (XAI) and a user interface (UI) tailored for medical professionals to address cardiovascular diseases (CVDs). We have developed a predictive model for heart disease with 96.07% accuracy, integrating factors like peak exercise ST segment slope, maximum heart rate, and exercise-induced angina. Our model prioritizes transparency, offering clear explanations for decisions to build trust in AI-driven healthcare. Emphasizing feature importance, it enables interpretable predictions, facilitating early detection and informed management. Our objective is two-fold: accurate risk identification and providing medical professionals with a user-friendly interface for transparent, reliable decision-making. This research advances healthcare by bridging AI with human understanding, enhancing outcomes in heart disease management.

Topics & Concepts