Determining cardiovascular disease prediction using data mining methods

Cardiovascular diseases (CVDs) are among the leading causes of death worldwide, affecting millions annually. According to the World Health Organization, early diagnosis can facilitate treatment and even prevention. However, traditional diagnostic methods remain expensive, time-consuming, and inefficient, discouraging regular medical follow-ups. In response, machine learning and artificial intelligence techniques offer the potential for accurate, fast, and cost-effective diagnoses. This study utilized a Kaggle dataset containing 1,190 samples from Cleveland, Hungary, Switzerland, and VA Long Beach (561 healthy and 629 with CVDs). The data were split into 80% for training and 20% for testing using Python. Various supervised learning algorithms, including K-Nearest Neighbors, Decision Trees, Random Forest, Gradient Boosting, Support Vector Machines, and Multilayer Perceptron, were trained and evaluated using 10-fold cross-validation. Gradient Boosting outperformed other models with 93% accuracy, precision, recall, and F1 score. These findings highlight the potential of AI-based techniques in facilitating early diagnosis and improving patient care in cardiovascular diseases.