Heart disease, responsible for 31% of global fatalities, remains a primary global cause of death. Recent advancements in medical support technology, employing data mining and machine learning, offer promising tool for prediction. This study focuses on machine learning algorithms to predict heart disease in Indian patients, utilizing tree-based techniques known for handling complex, nonlinear data. Preprocessing involves handling missing values and applying feature scaling techniques such as MinMaxScaler and StandardScaler. Feature selection is performed using Chi-squared and Analysis of Variance (ANOVA) tests for categorical and numerical features, respectively. Tree based algorithms namely Decision Tree, Random Forest, Gradient Boosting, Adaptive Boosting (AdaBoost), and Extreme Gradient Boosting (XGBoost) are evaluated using 5-fold cross-validation and grid search. Each model’s performance is assessed using metrics like accuracy, precision, recall, F1-score, and AUC-ROC (Area under the Receiver Operating Characteristic curve). Remarkably, the study identifies “age” and “gender” as non-correlating features and their removal improves model accuracy. The Random Forest model stands out as the top performer, achieving 99.5% accuracy, underlining its efficacy in heart disease prediction.