The mitral valve, one of the heart’s valves, is crucial for regulating blood flow throughout the body. Regurgitation is a sign of a valve problem that results in insufficiency. The valve partially closes, forcing blood from the lower ventricle back into the higher ventricle. Consequently, the heart will not pump enough blood to the body. This leads to cardiac failure as a result of the heart’s dysfunction. A heart attack could happen to some patients. Mitral valve failure can also be caused by other conditions, including high blood pressure, rheumatic fever, inactivity, etc. In order to start the right course of treatment for mitral valve dysfunction, an early diagnosis is essential. Several automated technological approaches, such as echocardiogram (ECHO), electrocardiogram (ECG), computed tomography (CT), and magnetic resonance imaging (MRI), can be used for detection. The image is examined with reference to the location of the mitral valve and the direction of blood pressure in order to detect the early warning symptoms of a heart attack. This paper shows a detailed analysis of various machine learning methods that are applied to classify images as normal or abnormal. There are several defined performance indicators, including accuracy, specificity, sensitivity, etc. Metric performance is improved through hyperparameter adjustment of the model. The techniques used are RandomizedSearchCV and GridSearchCV. Compared to the others, K-nearest neighbor had exceptionally high accuracy. The novelty approach used here is the way how the parameters are tuned such that the performance metrics can be increased, and thereby mitral valve failure can be determined accurately which outperforms existing one. One of the ancient Ayurvedic medicine Terminalia arjuna can be used to treat cardiovascular diseases. The rate of mitral valve failure is gradually reduced by blood pressure regulation.