Creating An Effective Machine Learning Model For Monitoring And Maintaining Cardiac Health

Heart disease (HD) is considered to be solitary of the greatest significant problems of the contemporary society as it concerns millions of people. Some are - breath shortness, physical weakness or weakness of the muscles, swollen feet and the like. This comes out clearly that researchers endeavor to establish effective methods to identify heart diseases in the primary stages because diagnostic methods currently in use are inconclusive and delay in diagnosis is rife. For the present project, we outline an optimal method for identifying the patients’ heart disease using the sophisticated ml method. The system involves the use of Voting Classifier to combine four classifiers - LR, K-NN, Random Forest, and Decision Trees. Through integration of the aspects of these different models the voting mechanism improves the degree of classification as well as time taken in execution. Classification accuracy and system efficiency are also optimized by using a feature choice algorithm. The performers employed performance measures like A-accuracy, P-pre-cision, R-re-call and F1-score to measure the impact of the classifiers. The final estimates are also exhibited through the Flask framework webpage makes it easy to use for the healthcare applications. Further, the system is equipped with patient-specific nutrition advice for patients with coronary disease for both, diagnostic and primary preventive purposes.