A Heart Disease Prediction in the Age of AI: A Comprehensive Review of ML and DL Approaches
Neha Patle1, Prof. Sarwesh Site 2
1 M.Tech Student, Department of Computer Science and Engineering All Saints College of Technology, Bhopal, India
Affiliated to Rajiv Gandhi Proudyogiki Vishwavidyalaya (RGPV) nehapatle88521@gmail.com
2 Associate Professor, Department of Computer Science and Engineering All Saints College of Technology, Bhopal, India
Affiliated to Rajiv Gandhi Proudyogiki Vishwavidyalaya (RGPV) er.sarwesh@gmail.com
Abstract - Heart disease prediction has become a critical challenge in modern healthcare, as cardiovascular diseases remain the leading cause of morbidity and mortality worldwide. Early and accurate prediction is inherently complex due to heterogeneous risk factors such as age, blood pressure, cholesterol, glucose levels, lifestyle habits, genetic predisposition, and comorbidities, which often lead to misclassification by conventional diagnostic systems. Earlier approaches based on machine learning with hand-crafted features and statistical models, or even deep learning architectures such as CNNs and LSTMs with static representations, have achieved limited success in addressing these complexities. The emergence of advanced AI techniques, including ensemble methods, attention-based models, and transformer architectures, has marked a paradigm shift in heart disease prediction by enabling multi-feature integration, temporal risk modeling, and improved generalization across diverse populations. This review presents a comprehensive examination of ML and DL methods for heart disease prediction, analyzing their architectures, training strategies, and advanced adaptations such as ensemble hybrids, explainable AI frameworks, and graph-based learning. A detailed discussion of benchmark datasets, evaluation metrics, and comparative performance across models is provided, offering insights into the strengths and weaknesses of these approaches relative to traditional baselines. Moreover, the review identifies persistent challenges such as dataset imbalance, feature variability, interpretability, limited real-world validation, and the urgent need for privacy-preserving predictive systems. Finally, it highlights emerging research directions, including multimodal integration (EHR, imaging, and genomics), explainable deep learning, federated learning for secure healthcare AI, and the use of large pre-trained models for clinical decision support. By consolidating recent advancements and open challenges, this study aims to serve as a foundational reference for researchers, practitioners, and policymakers working toward the development of robust, interpretable, and scalable AI systems for heart disease prediction.
Keywords: Heart Disease Prediction, Machine Learning, Deep Learning, Ensemble Models, Explainable AI, Healthcare Analytics, Clinical Decision Support.
