In the last few decades, heart disease has become much more common in people of all ages, so early detection became important. There are some things that make it harder to find heart disease, like diabetes, high blood pressure, an irregular heart rate, high cholesterol, and so on. To treat heart patients effectively, it is important to be able to properly diagnose heart disease before a heart attack happens. Machine learning-based noninvasive technology can swiftly and effectively identify heart disease patients. A machine-learning-based cardiovascular disease prediction system developed using heart disease datasets in the proposed research. Cross-validation used to evaluate machine learning, feature selection, and classifiers for accuracy and specificity. Here rapidly distinguish heart patients from healthy persons using technology. Receiver optimistic curves and area under the curves for each classifier were analyzed. Classifiers, feature selection algorithms, preprocessing methods, validation procedures, and performance measurements are covered in this study. A subset and the full set of features were used to test the suggested system's performance. Recall, F1 score, and false positive rate are compared. Decreases in the number of features utilized to classify affect accuracy and runtime. An expected machine-learning-based decision support system would help clinicians diagnose heart disease more accurately.

Heart disease Patients Machine learning Decision support system Clinicians diagnose.