Modeling of Artificial Intelligence Based Arrhythmia Detection and Classification Using Electrocardiography Signals on IoT Environment

Recently, a massive development in the control of IoT is witnessed which offers a lot in the healthcare approach as it enables constant patient observation and then there is a condition for advanced automated monitoring approaches for classification of heart attacks. An Electrocardiogram (ECG) can be extremely employed for estimating Cardiovascular Disease (CVD) that is outcome further explores heart conditions and detection of CVD. The process is fast, easy, and non-invasive. However, artificially classifying and identifying CVD can be complex, as manually inspecting the ECG signals is labour-intensive. This study presents Non-Dominated Sorting Genetic Algorithm with Optimal Graph Convolutional Network (NSGA-OGCN) model for arrhythmia detection. The intention of the NSGA-OGCN technique is the proper detection and classification of arrhythmia in the IoT atmosphere. Here, the NSGA-OGCN technique uses NSGA technique for optimal selection of feature subsets. For ECG signal detection, the GCN model is used. Finally, Adamax optimizer is utilized for the optimal hyperparameter tuning of the GCN model. The simulation results of the NSGA-OGCN methodology are examined on benchmark datasets. Comprehensive comparison analysis reported the improved accomplishment of the NSGA-OGCN methodology in terms of different measures.