Early Prediction of Cardiovascular Disease Using Convolutional Neural Networks: A Comparative Study with Traditional Machine Learning Models

Abstract

As a primary contributor to worldwide deaths, cardiovascular diseases necessitate timely detection to minimize potential health consequences. This research explores the application of Convolutional Neural Networks for the early prediction of heart disease, capitalizing on their capacity to automatically identify pertinent characteristics and recognize intricate structures within unprocessed medical information. The CNN model's effectiveness was assessed using the Cleveland Heart Disease dataset, resulting in a testing accuracy of 94.78%, surpassing the performance of conventional machine learning techniques, including Logistic Regression, Naïve Bayes, K-Nearest Neighbors, and Support Vector Machines. Moreover, the CNN model exhibited enhanced precision, recall, and F1-score values, further confirming its efficacy in detecting cardiovascular disease. Additionally, the model demonstrated outstanding performance on the Receiver Operating Characteristic curve, achieving an Area Under the Curve of 0.98. These results emphasize the potential of deep learning methodologies, particularly CNNs, to improve CVD prediction and aid healthcare practitioners in early diagnosis efforts. The combination of high accuracy and the model's capacity to generalize effectively to new data underscores the potential of CNNs to contribute to clinical decision support systems.