Recent Advances in AI-Based Heart Disease Prediction: A Review of Machine Learning and Deep Learning Approaches

Abstract

The current modality of diagnosis for heart diseases lacks the sensitivity and specificity characteristics necessary for early and accurate diagnosis of this time-sensitive illness. Artificial intelligence (AI), including machine learning (ML) and deep learning (DL), has demonstrated substantial potential in aiding cardiac health diagnosis and prediction. This review article provides a comprehensive review of ML and DL techniques used in predicting heart diseases. Additionally, it provides an overview of standard ML algorithms using K-Nearest Neighbour (KNN), Support Vector Machine (SVM), Decision Trees, Random Forests, and ensemble methods and advanced deep learning models such as Deep Neural Networks (DNN) and Convolutional Neural Networks (CNN) and hybrid architectures. The importance of data preprocessing, feature engineering, feature scaling, and data augmentation techniques that significantly influence its efficacy in a highly effective manner is brought to the fore in this review. Special attention is given to methods handling class imbalance, such as the Synthetic Minority Oversampling Technique (SMOTE), that improve the detection of minority classes and relevant medical datasets. In addition, optimisation strategies, activation functions, and regularisation techniques used in deep learning models are discussed to understand their role in improving the predictive accuracy and model generalisation. Algorithms and a hybrid algorithm for deep learning-related research show relatively better indications of performance in comparison to regression-based algorithms. To reach an even simpler precision, metrics must include accuracy, sensitivity, specificity, precision, F1-score, and AUC for model evaluation. Presented herein are also particularly creative reflections on further understanding of digital research and envisioning changes for future activities related to AI for the prediction of cardiac dysfunctions – because the fine-tuned discoveries will aid in general terms for evaluation and deployment as clinical feedback mechanisms.

How to cite this article:
Kumar R, Rai A K. Recent Advances in AI-Based Heart Disease Prediction: A Review of Machine Learning and Deep Learning Approaches. J Adv Res Appl Arti Intel Neural Netw 2026; 10(2): 1-9.