1 Department of Artificial Intelligence and Data Science, Rajalakshmi Institute of Technology, Chennai, Tamil Nadu, India
2 Department of Electrical and Electronics Engineering, New Horizon College of Engineering, Bengaluru, Karnataka, India
3 Department of Electrical and Electronics Engineering, Bharathiyar Institute of Engineering for Women, Deviyakurichi, India
4 Department of Electrical and Electronics Engineering, Sri Ranganathar Institute of Engineering and Technology, Athipalayam, Coimbatore, Tamil Nadu, India
5 Department of Electrical & Electronics Engineering, Sir M. Visvesvaraya Institute of Technology, Bengaluru, India
6 Department of Electrical and Electronics Engineering, Sri Eshwar College of Engineering, Coimbatore, Tamil Nadu, India
7 Department of Electrical & Electronics Engineering, Sona College of Technology, Salem, Tamil Nadu, India
8* Department of Physics, Government Arts College (Autonomous), Salem, Tamil Nadu, India (Corresponding Author). Email: sivaphotonics@gmail.com
Cardiovascular disease continues to be a significant issue for health all over the world and accounts for a notable percentage of deaths at an early age. Automated detection and prediction of heart disease through early detection can provide great assistance to medical staff in clinical decision-making. This research utilizes Machine Learning (ML) techniques to predict the existence of heart disease using a structured dataset of clinical descriptors. The study uses the Cleveland Heart Disease dataset from the UCI Machine Learning Repository, which contains 303 patient records and 13 clinical features. Initially, the dataset is pre-processed to address missing values, normalize features, and achieve class balance. Using feature selection techniques, such as the Chi-square statistical test and K-Best feature selection approach, the features are extracted. The research explores the performance of four ML algorithms, including Logistic Regression (LR), Decision Tree (DT), Support Vector Machine (SVM) and k-Nearest Neighbors (kNN). Overall model performance is assessed using performance metrics. The experimental results indicate that the use of feature selection techniques can vastly improve model efficiency and generalization. Among the models submitted, SVM with 94.8% accuracy indicated the best results and is an appropriate tool for Heart Disease.
Keywords: Medical diagnostics, Heart disease prediction, Machine learning, Feature selection, Cross-validation, Healthcare analytics, Predictive modeling, Clinical decision support
How to cite this article: Arthi A, Meenakshi Sundaram P, Surendiran S, Rajeshkumar K, Arulkumar T, PremKumar R, Yamuna KS, Sivakumar S. Interpretable Predictive Analytics for Early Cardiovascular Health Assessment. Int J Drug Deliv Technol. 2026;16(4): 317. DOI: 10.25258/ijddt.16.4.33
Source of support: Nil.
Conflict of interest: None