EUVIMEDEUVIMED entdecken
Crossref

Predicting Heart Disease Risk Using Hybrid CNN–NG Boost: An AI-Based Approach for Cardiovascular Health Prediction

S Mohanappriya, L Salomon Jeba Singh, JP Siva Prakash, V Selvapandeeswaran

International Journal of Engineering & Extended Technologies Research · 28.03.2026 · 08 (02)

DOI: 10.15662/ijeetr.2026.0802193 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Engineering & Extended Technologies Research, online erschienen am 28.03.2026. Verfasst von S Mohanappriya und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cardiovascular diseases (CVDs) remain the leading cause of mortality globally, necessitating the development of accurate and early risk prediction tools. While machine learning models have shown promise in this domain, they often face limitations in effectively capturing complex, non-linear patterns from high-dimensional clinical data and providing robust uncertainty estimates. This study proposes a novel hybrid AI framework, Hybrid CNN–NGBoost, which synergistically combines the feature learning prowess of a Convolutional Neural Network (CNN) with the probabilistic forecasting capabilities of Natural Gradient Boosting (NGBoost). The one-dimensional CNN acts as an intelligent feature extractor, automatically learning salient hierarchical patterns from raw, structured patient data. These enriched features are then fed into the NGBoost model, which not only performs the classification task but also outputs a full probability distribution for each prediction, quantifying the model's uncertainty. The proposed model was trained and evaluated on the publicly available Cleveland Heart Disease dataset. Its performance was benchmarked against several conventional machine learning algorithms, including Logistic Regression, Support Vector Machines, and standalone Random Forests. The Hybrid CNN–NG Boost model demonstrated superior predictive efficacy, achieving a peak accuracy of 96.3%, with precision, recall, and F1-score all exceeding 95%. Crucially, the model provides a calibrated measure of prediction confidence, a critical feature for clinical decision-making where understanding the "reliability" of a prediction is as important as the prediction itself. The results indicate that the Hybrid CNN–NGBoost framework is a highly effective and reliable tool for predicting heart disease risk. By offering high accuracy coupled with meaningful uncertainty quantification, this AI-based approach presents a significant advancement over traditional models and holds substantial potential for deployment as a decision-support system in clinical settings, ultimately aiding in proactive cardiovascular health management and personalized patient care

Zitieren

APA

Mohanappriya, S., Singh, L. S. J., Prakash, J. S., & Selvapandeeswaran, V. (2026). Predicting Heart Disease Risk Using Hybrid CNN–NG Boost: An AI-Based Approach for Cardiovascular Health Prediction. International Journal of Engineering & Extended Technologies Research, 08(02). https://doi.org/10.15662/ijeetr.2026.0802193

Vancouver

Mohanappriya S, Singh LSJ, Prakash JS, Selvapandeeswaran V. Predicting Heart Disease Risk Using Hybrid CNN–NG Boost: An AI-Based Approach for Cardiovascular Health Prediction. International Journal of Engineering & Extended Technologies Research. 2026;08(02). doi:10.15662/ijeetr.2026.0802193

BibTeX
@article{Mohanappriya2026,
  author = {Mohanappriya, S and Singh, L Salomon Jeba and Prakash, JP Siva and Selvapandeeswaran, V},
  title = {Predicting Heart Disease Risk Using Hybrid CNN–NG Boost: An AI-Based Approach for Cardiovascular Health Prediction},
  journal = {International Journal of Engineering & Extended Technologies Research},
  year = {2026},
  volume = {08},
  number = {02},
  doi = {10.15662/ijeetr.2026.0802193}
}
Kostenlos starten →