A Survey on Hybrid Deep Learning and Neural Fuzzy Inference Systems for Early Coronary Heart Disease Detection

coronary heart disease (CHD) continues to be a primary cause of global mortality, highlighting the critical necessity for precise and early detection techniques to enable prompt management and prevention. Recent breakthroughs in ML and DL have demonstrated potential in improving diagnosis accuracy; yet, challenges remain regarding interpretability, computational complexity, and the management of ambiguous or unclear medical data. This survey examines advanced methodologies and investigates the possibility of hybrid frameworks that combine deep learning models with neural fuzzy inference systems (NFIS) for the identification and prevention of CHD. Hybrid techniques, which integrate the feature extraction and learning capabilities of deep learning with the interpretability and uncertainty management of neuro-fuzzy inference systems, provide a robust framework for enhancing early coronary heart disease diagnosis and risk evaluation. We offer an extensive comparison of modern machine learning, deep learning, and hybrid models, evaluating their performance across multiple measures, including accuracy, sensitivity, and computing demands. This paper examines upcoming topics such as transfer learning, multi-modal data integration, and explainable AI, emphasising the prospective applications of these systems in clinical environments. Our research indicates that hybrid DL-NFIS models possess considerable potential for improving CHD identification and, thus, augmenting patient outcomes in preventive healthcare.