EUVIMEDEUVIMED entdecken
Crossref

Hybrid Multi-Backbone Feature Extraction with Grey Wolf Optimization and Ensemble Learning for Explainable Brain Stroke Detection from CT Images

Sriteja Paka, Antham Rajini, Mallu Sathwika, Gnanadeep Gumpula, Mangu Akhila

Journal of Computing in Education, Sports and Health · 30.04.2026 · 3 (1) · 11-19

DOI: 10.69996/jcesh.2026002 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Computing in Education, Sports and Health, online erschienen am 30.04.2026. Verfasst von Sriteja Paka und 4 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cerebrovascular diseases such as stroke are among the most common causes of the death and the disability worldwide and are preventable and treatable. Early detection of strokes and their rapid intervention play an important role in reducing the burden of disease and improving clinical outcomes. Early stroke detection from CT images remains clinically challenging due to subtle intensity variations and complex anatomy. This paper presents a hybrid framework integrating multi-backbone feature extraction (EfficientNetV2-S, EfficientNetV2-M, DenseNet201) gen- erating 4480 features, Grey Wolf Optimizer (GWO) reducing features to 594 (86.7 percent reduction), dual classifiers (regularized ANN: 96.93 percent, BiLSTM: 95.90 percent), and validation-weighted ensemble (0.502/0.498) achieving 95.90 percent accuracy. SHAP-based explainability provides clinical interpretability. Ablation studies confirm GWO improves gen- eralization by 8.26 percent. The framework demonstrates that strategic integration of multi- representational features, metaheuristic optimization, and structured ensemble learning offers a robust, interpretable, and computationally feasible solution for automated stroke detection.

Zitieren

APA

Paka, S., Rajini, A., Sathwika, M., Gumpula, G., & Akhila, M. (2026). Hybrid Multi-Backbone Feature Extraction with Grey Wolf Optimization and Ensemble Learning for Explainable Brain Stroke Detection from CT Images. Journal of Computing in Education, Sports and Health, 3(1), 11-19. https://doi.org/10.69996/jcesh.2026002

Vancouver

Paka S, Rajini A, Sathwika M, Gumpula G, Akhila M. Hybrid Multi-Backbone Feature Extraction with Grey Wolf Optimization and Ensemble Learning for Explainable Brain Stroke Detection from CT Images. Journal of Computing in Education, Sports and Health. 2026;3(1):11-19. doi:10.69996/jcesh.2026002

BibTeX
@article{Paka2026,
  author = {Paka, Sriteja and Rajini, Antham and Sathwika, Mallu and Gumpula, Gnanadeep and Akhila, Mangu},
  title = {Hybrid Multi-Backbone Feature Extraction with Grey Wolf Optimization and Ensemble Learning for Explainable Brain Stroke Detection from CT Images},
  journal = {Journal of Computing in Education, Sports and Health},
  year = {2026},
  volume = {3},
  number = {1},
  pages = {11--19},
  doi = {10.69996/jcesh.2026002}
}
Kostenlos starten →