Machine learning approaches for cardiovascular disease prediction: A review - 14/06/25

Doi : 10.1016/j.acvd.2025.04.055

Siming Wan ^a,^b,^c, Feng Wan ^c, Xi-jian Dai ^a,^b,⁎
^a Department of Radiology, The Second Affiliated Hospital, Jiangxi Medical College, Nanchang University, 330006 Nanchang, China
^b Jiangxi Provincial Key Laboratory of Intelligent Medical Imaging, 330006 Nanchang, China
^c Department of Electrical and Computer Engineering, Faculty of Science and Technology, University of Macau, Taipa, Macau, China

^⁎Corresponding author. Department of Radiology, The Second Affiliated Hospital, Jiangxi Medical College, Nanchang University, Minde Road No. 1, 330006 Nanchang (Jiangxi Province), China.Department of Radiology, The Second Affiliated Hospital, Jiangxi Medical College, Nanchang UniversityMinde Road No. 1Nanchang (Jiangxi Province)330006China

Sous presse. Épreuves corrigées par l'auteur. Disponible en ligne depuis le Saturday 14 June 2025
Cet article a été publié dans un numéro de la revue, cliquez ici pour y accéder

Graphical abstract

Le texte complet de cet article est disponible en PDF.

Highlights

•	Comparison of machine learning and traditional methods in CVD prediction.
•	Evaluation metrics: accuracy, precision, F1-score, sensitivity, specificity, AUC.
•	Data collection and preprocessing: importance, sources, preprocessing techniques.
•	Cross-validation: k-fold cross-validation.
•	Current challenges: regulatory and ethical issues, deep learning interpretability.
•	Future research: data preprocessing impact, data privacy, IoT integration.

Le texte complet de cet article est disponible en PDF.

Abstract

Cardiovascular disease is a leading cause of death worldwide and is associated with significant morbidity and mortality. The use of artificial intelligence techniques, particularly machine learning algorithms, has emerged as a transformative approach for enhancing early diagnostic accuracy of disease compared with conventional diagnostic methods. This systematic review examines three core aspects: (1) comparative analysis of current machine learning algorithms in early diagnosis of cardiovascular disease, (2) operational frameworks for clinical implementation, and (3) critical evaluation of regulatory compliance and ethical implications. It summarizes recent advancements in machine learning-based heart disease prediction, outlines a typical workflow for applying machine learning in clinical settings, and discusses the regulatory and ethical challenges associated with its implementation. Finally, this review explores potential directions for future research in this rapidly evolving field.

Le texte complet de cet article est disponible en PDF.

Keywords : Machine learning, Cardiovascular disease, Classification, Prediction, Feature selection

Abbreviations : AI, AUC, CVD, EHR, ICD-11, IoT, ML, ROC, SMOTE