SBB-Chi²-A²: stacking of bagging-boosting with the blend Chi-square for effective prediction of aortic aneurysm using biomarker profiling

作     者：Jena, Sanjuktarani Brahma, Biswajit Khan, Zabiha Jyothi, G. Arunachalam, P. Aggarwal, Saurabh 

作者机构：Department of Computer Science and Engineering Sardar Patel Institute of Technology (SPIT) Andheri West Mumbai India McKesson Corporation 32559 Lake Bridgeport St Fremont 94555 CA United States Nitte Meenakshi Institute of Technology Bengaluru India Department of Pharmaceutical Engineering B V Raju Institute Of Technology Telangana Narsapur India Department of Biomedical Engineering Vel Tech Rangarajan Dr.Sagunthala R & D Institute of Science and Technology Tamil Nadu Chennai India San Jose State University San Jose United States 

出 版 物：《International Journal of Information Technology (Singapore)》 (Int. J. Inf. Technol.)

年 卷 期：2025年

页      面：1-8页

主　　题：AdaBoost Aortic Aneurysm (A²) disease Bagging Biomarker profiling Boosting Chi-square (Chi²) 

摘      要：Aortic Aneurysm (A2) remains a leading cause of morbidity and mortality worldwide, necessitating innovative approaches for accurate prediction and early intervention. This study proposes a novel ensemble learning framework that integrates bagging and boosting techniques through stacking, combined with Chi-square feature selection, to enhance the prediction of A2 condition using biomarker profiling. The stacking method leverages the strengths of individual models, including Decision Trees (DT), Random Forests (RF), and AdaBoost, GBoost, and XGBoost, to create a robust meta-model. The experimental evaluations are then conducted on Aorta vessel Tree (AVT) dataset taken from public repository. Chi-square (Chi2) feature selection is utilized to identify the most significant biomarkers, reducing dimensionality and ensuring that only the most relevant features contribute to the model. This preprocessing step enhances the model’s interpretability and performance by focusing on the critical factors associated with Aortic Aneurysm. The proposed framework is evaluated using key performance metrics, demonstrating superior predictive performance compared to conventional individual-model approaches. © Bharati Vidyapeeth s Institute of Computer Applications and Management 2025.