Enhancing Prognosis Accuracy for Ischemic Cardiovascular Disease Using K Nearest Neighbor Algorithm: A Robust Approach

Abstract

Ischemic Cardiovascular diseases are one of the deadliest diseases in the world. However, the mortality rate can be significantly reduced if we can detect the disease precisely and effectively. Machine Learning (ML) models offer substantial assistance to individuals requiring early treatment and disease detection in the realm of cardiovascular health. In response to this critical need, this study developed a robust system to predict ischemic disease accurately using ML-based algorithms. The dataset obtained from Kaggle encompasses a comprehensive collection of over 918 observations, encompassing 12 essential features crucial for predicting ischemic disease. In contrast, much-existing research relies primarily on datasets comprising only 303 instances from the UCI repository. Six ML-based algorithms, including K Nearest Neighbors (KNN), Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), Gaussian Naïve Bayes (GNB), and Decision Trees (DT), are trained on the ischemic heart data. The effectiveness of the proposed methodologies is meticulously evaluated and benchmarked against cutting-edge techniques, employing a range of performance criteria. The empirical findings manifest that the KNN classifier produced optimized results with 91.8% accuracy, 91.4% recall, 91.9% F1 score, 92.5% precision, and AUC of 90.27%.