Cardiovascular Disease (CVD) is a major reason for general mortality rate. According to WHO it is the leading cause of death worldwide, resulting in 17.9 million fatalities annually, or roughly 31% of the total mortality worldwide. Devices like ECG, echocardiogram, Holter monitor, cardiac MRI, etc. are used for detection of heart disease in hospitals. Usually, choosing the best features is challenging. In order to solve this problem, ideal Feature Selection (FS)-based Machine Learning (ML) techniques are suggested for early prediction. Using ML classifiers, the system, which consists of components for processing, storing, and gathering data, predicts patients heart problems. Least Absolute Shrinkage and Selection Operator (LASSO), SHapley Additive exPlanations (SHAP), Analysis of Variance (ANOVA), and Minimum Redundancy Maximum Relevance (mRMR) techniques are applied for feature extraction. Further, we used SHAP with Support Vector Machine (SVM), Logistic Regression (LR), Random Forest (RF), Extreme Gradient Boosting (XGBOOST), and Decision Tree (DT) for prediction. mRMR, LASSO, ANOVA are applied with Random Forest Classifier (RFC), Gradient Boost Classifier (GBC), Extra Tree Classifier (ETC), and Logistic Regression Classifier (LRC) for prediction. SHAP with SVM achieves highest accuracy with 86%. ANOVA with LRC achieves 85%. The suggested approach has the power to significantly reduce mortality from CVD and improve patient care, improving the lives of those who are impacted.