IoT Model for Heart Disease Detection Using Machine Learning (ML) Techniques

From the last decade, a tremendous spotlight is on the giving quality medical services because of the exponentially growing of life threatening illnesses of the patients. There are numerous components that influence the health condition of each person and a few illnesses are more dangerous and cause death of the patient. And in the present age, the most common reason of death is heart disease. This research work presents the IoT based system for heart disease detection using the Machine Learning (ML) technique. It consists a novel preprocessing stage that provides more accurate classification of the ECG signal. Also, this novel preprocessing method removes the noise effectively from the raw ECG data. The classification performance was evaluated using the various classifiers such as KNN, Naïve Bayes and Decision tree that detects the normal and abnormal heart-beat rhythms. With the obtained results, we have observed that the preprocessing has improved the classification performance. This technique further proves that the decision tree has good performance over the KNN and Naïve Bayes with respect to the accuracy, sensitivity and precision.