Heart disease, a leading cause of global mortality, can significantly impact overall health. Timely prediction and identification of key risk factors are essential yet challenging. This study employs Machine Learning and Explainable Artificial Intelligence (XAI) to predict and identify major contributors to heart disease. Using a publicly available cardiovascular disease (CVD) dataset from Kaggle, consisting of 1319 samples with eight variables—age, gender, heart rate, systolic and diastolic blood pressure, blood sugar, CK-MB, and Test-Troponin—we trained and evaluated models for heart attack prediction. Supervised learning techniques, including SVM, Decision Tree, and Random Forest, were compared. The LIME technique was used to elucidate the influence of factors in different test cases. The study aimed to enhance classification and detection of key features leading to heart attacks. Model performance was assessed using metrics such as confusion matrix, accuracy, precision, recall, and F1 score. In the medical context, minimizing false negatives (missed actual positive cases) is critical. Random Forest emerged as the most effective model for reducing false negatives, making it a promising tool for heart disease prediction.