Predicting Cardiovascular Disease in Diabetic Patients Using Deep Neural Networks
PMRP Journal of Research – Part A: Computational Engineering · 10.05.2026 · 2 (1) · 13-27
DOI: 10.65383/jra/2026.2.1.2 ↗Kurz gefasst
Fachartikel in der Zeitschrift PMRP Journal of Research – Part A: Computational Engineering, online erschienen am 10.05.2026. Verfasst von C Rukumani Khandhan und 5 weiteren Autor:innen.
Abstract (Originalsprache)
The health problem of cardiovascular disease is a serious health concern across the globe, especially in patients with diabetes, as they are more prone to the metabolic effects of failure to control glucose. Diabetic patients should have their cardiac conditions detected early to limit complications and enhance patient outcomes. Although clinical records contain high volumes of patient data, in the majority of healthcare organizations, transforming patient data into valuable clinical data is a challenging practice. To eliminate this problem, a predictive model is designed based on deep learning models to determine cardiovascular disease in diabetic patients through the UCI Heart Disease dataset. A multilayer feedforward neural network was used to model the complex associations of various clinical measurements. It is a four-layer model, where the first layer consists of 128 neurons, the second layer comprises 64 neurons, the third layer consists of 64 neurons and the fourth layer has 32 neurons. The use of batch Normalization and dropout at the end of every hidden layer is employed to ensure effective learning and minimize overfitting. The hidden layers use the activation function, namely Rectified Linear Unit (ReLU), whereas the sigmoid function is used in output layer to perform binary classification. The experimental evidence shows that the model proposed has an accuracy of 91.02%, denoting good predictive activity in detecting heart disease in diabetic patients. These results indicate that deep learning algorithms have the potential to support efficient clinical decision-support systems.
Zitieren
APA
Khandhan, C. R., Gothai, E., Arumugam, M., Deepa, S., Jamunadevi, C., & Raja, R. K. (2026). Predicting Cardiovascular Disease in Diabetic Patients Using Deep Neural Networks. PMRP Journal of Research – Part A: Computational Engineering, 2(1), 13-27. https://doi.org/10.65383/jra/2026.2.1.2
Vancouver
Khandhan CR, Gothai E, Arumugam M, Deepa S, Jamunadevi C, Raja RK. Predicting Cardiovascular Disease in Diabetic Patients Using Deep Neural Networks. PMRP Journal of Research – Part A: Computational Engineering. 2026;2(1):13-27. doi:10.65383/jra/2026.2.1.2
BibTeX
@article{Khandhan2026,
author = {Khandhan, C Rukumani and Gothai, E and Arumugam, M and Deepa, S and Jamunadevi, C and Raja, R Kisor},
title = {Predicting Cardiovascular Disease in Diabetic Patients Using Deep Neural Networks},
journal = {PMRP Journal of Research – Part A: Computational Engineering},
year = {2026},
volume = {2},
number = {1},
pages = {13--27},
doi = {10.65383/jra/2026.2.1.2}
}
EUVIMED