DETECTION OF CARDIOVASCULAR DISEASES THROUGH MACHINE LEARNING AND DEEP LEARNING APPROACHES

Abstract

Cardiovascular diseases (CVDs) claim millions of lives each year and continue to pose a major threat to global health. Machine learning (ML) techniques are promising to enhance diagnostic accuracy, which is essential for early detection and effective treatment. In this study, various ML algorithms are explored for identifying CVDs using datasets containing patient demographics and key health indicators. Recent research highlights advanced methods such as hybrid genetic algorithms for diagnosing atherosclerotic heart disease and random search algorithms for predicting heart failure. The effectiveness of models like Decision Trees, Naïve Bayes, and Random Forests in classifying cardiac conditions has also been examined.

This paper proposes and evaluates a novel neural network architecture for cardiovascular disease prediction. A detailed performance analysis—using metrics such as accuracy, precision, recall, and F1-score—demonstrates the capability and potential of ML models in improving cardiovascular healthcare. The results are promising, with the proposed prediction model achieving a classification accuracy of 74%, indicating its potential to support physicians in timely interventions and improved patient outcomes.