Citation

Afroz, Sonia and Islam, Md Munjurul and Sharma, Nikunj and Rawat, Kanika and Saha, Tuhalika and Das, Rana and Mim, Sheikh Urvana Akter and Alam, Mst Sanjida and Ali, Md. Akkas (2025) Predicting Heart Disease with Machine Learning: A Comparative Analysis of Multiple Models. In: 2025 5th International Conference on Emerging Research in Electronics, Computer Science and Technology (ICERECT), 12-13 September 2025, MANDYA, India.

Text 8.pdf - Published Version Restricted to Repository staff only Download (2MB)

Abstract

While cardiovascular disease (CVD) remains a leading cause of global mortality, accurate predictive models that enable timely diagnosis and intervention are critical. This paper provides an overview of a selection of ML models used for cardiac disease prediction through a comparative study concerning the performance, as well as explainability. We evaluate several algorithms LR, RF, SVM, and DT using a well-known dataset publicly available, which includes clinical measurements. The models are assessed using standard performance metrics. The deep learning models that exceed baseline models have achieved as accuracy (85%, 81%, 81%, 80%), sensitivity (recall) and specificity (80%, 80%, 80%, 77%), precision (92%, 86%, 86%, 85%), F1-Score (86%, 83%, 83%, 81%), MSE (0.15, 0.19, 0.19, 0.20), and ROC-AUC (90%, 87%, 89%, 80%) for LR, RF, SVM, and DT models, respectively. LR achieved the highest accuracy of 85% and ROC-AUC score of 90%, proving to be very efficient in predicting nonlinear patterns and reducing misclassification errors. These results will highlight the positive and negative aspects of each model, where Random Forest and Neural Networks are superior in terms of accuracy and ROC-AUC, whilst LR yields simplicity and interpretability. The work dissects the complex interplay of model complexity and interpretability and highlights which methods are best for predicting cardiac disease in clinical settings. The work will eventually transform how AI-enabled diagnostic devices evolve and set the bar for further research and real-world adoption in health matters.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Heart Disease, Predictive Models, AUC, Healthcare, Early Diagnosis, Clinical Features
Subjects:	Q Science > QP Physiology > QP351 Neurophysiology and Neuropsychology
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	20 Apr 2026 02:02
Last Modified:	20 Apr 2026 02:02
URII:	http://shdl.mmu.edu.my/id/eprint/15748

Downloads

Edit (login required)