An Explainable Hybrid Intelligent System for Prediction of Cardiovascular Disease

Abstract

Cardiovascular disease is one the major cause of death around the world. Even while medical science continues to assist efforts to save lives, qualified medical professionals are still in limited. Accurate diagnosis at the right time is crucial in cardiovascular disease cases, as patients might live a long life with the right medical care. Machine learning and artificial intelligence have a significant impact on the early and precise prediction of cardiovascular disease. In this paper a machine learning based model for cardiovascular disease prediction has been proposed applying Logistics Regression, Naïve Bayes, K-Nearest Neighbor, Support Vector machine, Kernel SVM, Decision Tree classifier, Random Forest and Artificial Neural network with model explanation using Explainable AI. Based on the precision, specificity, and sensitivity scores of each method, the most effective one has been selected. Local Interpretable Model Agnostic Explanation (LIME) and Shapely Value (SHAP) have been used for model explanation.