A novel intelligent deep optimized framework for heart disease prediction and classification using ECG signals

The advancement in medical diagnosis approaches increases the demand for effective disease prediction and classification system. Although, various machine learning (ML) based disease classification techniques were developed, they face severe issues. Hence, a novel optimized framework named as Wolf based Generative Adversarial System (WbGAS) system was designed to predict and specify the heart disease using Electrocardiogram (ECG) database. The collected dataset contains three classes namely Normal Sinus Rhythm (NSR), Arrhythmia (ARR), and Congestive Heart Failure (CHF). The dataset is initialized and trained using the proposed (WbGAS) approach to predict the normal and abnormal signals present in dataset. In addition, the integration of wolf fitness function in the presented approach provides finest prediction rate. Moreover, the type of heart disease is specified based on the trained features. Also, a case study was presented with three different cases to explain the functioning of designed (WbGAS) approach. The designed model is implemented in MATLAB software and then, the performance of the system is determined as specificity, recall, accuracy, and precision value. At the end, to verify the results of the developed technique a comparative assessment was performed by comparing the outcomes of presented approach with existing ML based approaches.