Enhancing Heart Disease Classification: Performance Evaluation of the Machine Learning Model

Performance Evaluation of the Machine Learning Model

Abstract

Heart disease, mainly caused by coronary artery disease, poses a significant health challenge by impeding oxygen-rich blood flow to the heart muscle. Addressing this, our study utilizes a dataset of 1,026 cases from Kaggle, focusing on the application of MultiLayer Perceptron (MLP) in diagnosing heart conditions. This dataset, rich in attributes indicative of heart disease, comprises 14 variables with the class variable determined by 13 attributes acting as the focal point for disease classification. Our methodology involved applying various machine learning algorithms, with a particular emphasis on the MLP model, to categorize cases into diseased or non-diseased. The MLP model demonstrated superior performance, achieving an accuracy of 92.39% (AUC = 0.923), and outperforming several comparative techniques in terms of precision, recall, and mean squared error (MSE). The analysis of the dataset under this model revealed insightful patterns and trends, underscoring the effectiveness of MLP in heart disease diagnosis and potentially guiding future research in medical diagnostics.