2026-08-21T02:51:06.952Z http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai

ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai:3:12152 2026-08-19T21:42:58Z 3

Cardiovascular Diseases Risk Prediction Using XGBoost Classifier Optimized by Adapted Firefly Algorithm, chapter in LNNS Lecture Notes in Networks and Systems: ICAIN 2025: International Conference on Artificial Intelligence and Networks, Springer, volume 1930 2026 http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai/record/3/12152 https://link.springer.com/chapter/10.1007/978-3-032-23317-2_39 P. Dabic J. Petrovic T. Zivkovic M. Zivkovic B. Radomirovic S. Anetic I. Kosta V. Zeljkovic N. Bacanin Providing a precise evaluation of patient risk pertaining to different cardiovascular conditions is considered critical for prioritizing patients suitably for treatment. In this regard, the current research work proposes a solution for estimating patient-risk factors for cardiovascular conditions within healthcare units, based on a carefully optimized machine learning approach. The classification problem in this work used the XGBoost algorithm, which had hyperparameters optimized by a modified version of the firefly algorithm. The approach showed high accuracy in determining patient risks concerning heart conditions, with models delivering approximately 91.68% accuracy rates. The results of this research, therefore, indicate the feasibility of the technique as a useful tool with potential to facilitate faster responses in healthcare, particularly with respect to resource handling within healthcare units. bookPart Springer, Cham 504 519 10.1007/978-3-032-23317-2_39 LNNS Lecture Notes in Networks and Systems: ICAIN 2025: Proceedings of International Conference on Artificial Intelligence and Networks, volume 1930