Smart Healthcare: IoT-Enabled ECG Analysis and ML-Based Severity Prediction for Heart Disease

The work describes a smart design and implementation of an Electrocardiogram (ECG) Heart Rate Monitoring System. The system uses AD8232 ECG module, a high-precision heart rate monitoring device, and also has a machine learning algorithms to predict the probability of the occurrence of possible heart diseases. The system basically focuses on the ability to perform real-time ECG monitoring. The function is crucial because the opportunity to detect any kind of abnormalities in the electrical activity of the heart arises. If the abnormalities are detected early, then it becomes much easier to have timely medical treatment, and the patients can be immensely improved thereafter. The main component is AD8232 module for the system. The ECG signals are captured using the module, further these signals are processed and analysed using the most advanced machine learning techniques. The system's machine learning component employs a number of techniques, including Random Forest, Decision Trees, Support Vector Machines, and Logistic Regression. All these algorithms work together in order to classify the ECG signal and in order to detect any sign of a heart disease. Random Forest algorithm is much more suited than other models which can be concluded based on its better accuracy result. The system is portable, compact, cost effective and user-friendly; therefore it can be used both in the clinical and personal domain.