Heart Disease Prediction Using Ensemble Methods, Genetic Algorithms, and Data Augmentation: A Preliminary Study
Journal of Robotics and Control (JRC) · 06.05.2025 · 6 (3) · 1092-1105
DOI: 10.18196/jrc.v6i3.25144 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Robotics and Control (JRC), online erschienen am 06.05.2025. Verfasst von Deepali Yewale und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Statistically speaking, heart disease (HD) accounted for 1 in 5 fatalities in 2022, demanding affordable and accurate diagnosis. Traditional methods of prediction are accurate but expensive, creating a demand for sophisticated and efficient technologies. One of the most popular methods that researchers employ to forecast diseases is machine learning (ML). The goal of this effort is to improve HD prognosis accuracy through the use of ensemble approaches, specifically Random Forest (RF), XGBoost, Voting, and Stacking methods, which improve prediction accuracy by combining multiple models to capture complex patterns. Genetic algorithms (GA) are used to prioritize features. Incorporating data balancing, outlier removal techniques, and data augmentation, creates a model that delivers performance comparable to state-of-the-art research. Methods like random oversampling address data imbalance, while an isolation forest is employed to identify anomalies. To increase the dataset size and improve model performance, random noise is added after anomaly removal. Performed the cross-validation and robustness checks to assess the model's performance on both augmented and non-augmented datasets, ensuring that the inclusion of random noise did not excessively affect generalizability or result in overfitting. The proposed model’s effectiveness is evaluated using various performance metrics. Achieving 99.36% accuracy, 98% sensitivity, 100% specificity, 100% PPV, 97% NPV, 0.99 F-score, and an AUC of 1, the methodology shows great promise as a cost-effective, accurate, and highly efficient diagnostic tool for heart disease. The model's short training time and high performance suggest its potential for practical implementation in clinical settings, offering a reliable and affordable solution for early heart disease detection.
Zitieren
APA
Yewale, D., Patil, S., Date, A. R., & Nanthaamornphong, A. (2025). Heart Disease Prediction Using Ensemble Methods, Genetic Algorithms, and Data Augmentation: A Preliminary Study. Journal of Robotics and Control (JRC), 6(3), 1092-1105. https://doi.org/10.18196/jrc.v6i3.25144
Vancouver
Yewale D, Patil S, Date AR, Nanthaamornphong A. Heart Disease Prediction Using Ensemble Methods, Genetic Algorithms, and Data Augmentation: A Preliminary Study. Journal of Robotics and Control (JRC). 2025;6(3):1092-1105. doi:10.18196/jrc.v6i3.25144
BibTeX
@article{Yewale2025,
author = {Yewale, Deepali and Patil, Swati and Date, Archana Rajesh and Nanthaamornphong, Aziz},
title = {Heart Disease Prediction Using Ensemble Methods, Genetic Algorithms, and Data Augmentation: A Preliminary Study},
journal = {Journal of Robotics and Control (JRC)},
year = {2025},
volume = {6},
number = {3},
pages = {1092--1105},
doi = {10.18196/jrc.v6i3.25144}
}
EUVIMED