EUVIMEDEUVIMED entdecken
Crossref

Secure Heart Disease Status Classification Using Machine Learning and Blockchain-Based EHR Integrity Verification

Haya H. Alsharif, Sabah M. Alzahrani

International Journal of Advanced Computer Science and Applications · 2026 · 17 (7)

DOI: 10.14569/ijacsa.2026.0170757 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Advanced Computer Science and Applications, 2026. Verfasst von Haya H. Alsharif und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Heart disease remains one of the leading causes of mortality worldwide, motivating the development of accurate and scalable screening-support tools. This study presents an integrated framework for heart-disease status classification using the Heart 2020 Cleaned dataset, derived from the CDC Behavioral Risk Factor Surveillance System (BRFSS), a telephone-based survey of self-reported health information rather than clinical or electronic-health-record data. The task is therefore to classify a respondent’s self-reported heart-disease status, not to predict future disease onset. The pipeline includes categorical encoding, feature scaling, recursive feature elimination, and class-imbalance handling using BorderlineSMOTE, SMOTETomek, and SMOTEENN. Five classifiers were evaluated: Random Forest, XGBoost, LightGBM, k-Nearest Neighbours, and a feed-forward deep neural network. Sampling strategy affected reported performance. Under the methodologically valid post-split setting, in which resampling is applied only to the training data, the models achieved an AUC of approximately 0.85, comparable to prior work on the same dataset. Pre-split global resampling produced much higher scores, with Random Forest reaching 97.14% accuracy, an F1-score of 0.9735, and an AUC of 0.9965 under SMOTEENN. However, these results are optimistically biased because resampling before splitting leaks synthetic information into the test set. The system also includes a blockchain-based integrity layer that stores full records in the application database while anchoring SHA-256 digests and verification metadata in an Ethereum-compatible environment. Overall, this prototype demonstrates the practical integration of established machine-learning classification and blockchain-supported integrity-verification components.

Zitieren

APA

Alsharif, H. H., & Alzahrani, S. M. (2026). Secure Heart Disease Status Classification Using Machine Learning and Blockchain-Based EHR Integrity Verification. International Journal of Advanced Computer Science and Applications, 17(7). https://doi.org/10.14569/ijacsa.2026.0170757

Vancouver

Alsharif HH, Alzahrani SM. Secure Heart Disease Status Classification Using Machine Learning and Blockchain-Based EHR Integrity Verification. International Journal of Advanced Computer Science and Applications. 2026;17(7). doi:10.14569/ijacsa.2026.0170757

BibTeX
@article{Alsharif2026,
  author = {Alsharif, Haya H. and Alzahrani, Sabah M.},
  title = {Secure Heart Disease Status Classification Using Machine Learning and Blockchain-Based EHR Integrity Verification},
  journal = {International Journal of Advanced Computer Science and Applications},
  year = {2026},
  volume = {17},
  number = {7},
  doi = {10.14569/ijacsa.2026.0170757}
}
Kostenlos starten →