CoroNet: Predicting Coronary Artery Disease using Deep Convolutional Neural Networks and Ensemble Model for Enhanced Diagnostic Precision

The coronary artery disease (CAD) is a cardiovascular ailment marked by the constriction or obstruction of the coronary arteries, hindering the circulation of blood to the heart muscle and potentially causing chest pain, heart attacks, or other complications in the cardiac system. Forecasting Coronary Artery Disease with deep learning encounters difficulties including scarce and unbalanced data due to privacy issues and the infrequency of certain conditions, alongside the imperative for transparency and comprehensibility in the decision-making of robust yet inscrutable models such as Deep Convolutional Neural Networks (CNNs). In mitigating the constraints posed by limited and imbalanced data in the prediction of CAD, sophisticated deep learning approaches, including the utilization of Ensembles of Deep CNNs and implementing Transfer Learning with Pre-trained Models, are utilized to improve the resilience and adaptability of the models. The suggested system, utilizing advanced Deep CNN techniques, provides increased predictive precision by utilizing ensemble models and transfer learning, thereby improving resilience through the use of pre-trained models for optimal feature extraction. This study indicates that the Deep CNN technique attained an 81.97% accuracy, while Ensembles of Deep CNNs showcased a marginally superior accuracy of 83.61.% in predicting coronary artery disease, emphasizing the subtle distinctions in their predictive capabilities.