Dynamic Ensemble Learning with Gradient-Weighted Class Activation Mapping for Enhanced Gastrointestinal Disease Classification
Electronics · 14.01.2025 · 14 (2) · 305
DOI: 10.3390/electronics14020305 ↗Kurz gefasst
Fachartikel in der Zeitschrift Electronics, online erschienen am 14.01.2025. Verfasst von Chih Mao Tsai und 1 weiteren Autor:innen. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science and Technology Council (NSTC), Taiwan, Republic of China.
Abstract (Originalsprache)
Gastrointestinal (GI) disease classification through endoscopic images is a critical yet challenging task, due to inter-class variability and subtle feature overlaps. This study introduces a novel ensemble learning framework that combines case-specific dynamic weighting with Gradient-weighted Class Activation Mapping (Grad-CAM) to enhance both accuracy and interpretability. Three state-of-the-art convolutional neural networks (DenseNet201, InceptionV3, and VGG19) were fine-tuned on the Kvasir v2 dataset and integrated using optimized weights (0.4, 0.4, and 0.2, respectively). The ensemble achieved an accuracy of 0.91, outperforming individual models, particularly in complex classes such as Esophagitis and Normal-Z-Line. Grad-CAM visualizations confirmed the ensemble’s focus on clinically relevant features, highlighting its potential to improve diagnostic interpretability. While the dynamic ensemble approach significantly enhanced classification performance, further refinement is needed to address subtle and ambiguous cases. These results underscore the promise of dynamic ensemble learning as an explainable and clinically applicable tool in medical imaging.
Förderung
- National Science and Technology Council (NSTC), Taiwan, Republic of China
Zitieren
APA
Tsai, C. M., & Lee, J. (2025). Dynamic Ensemble Learning with Gradient-Weighted Class Activation Mapping for Enhanced Gastrointestinal Disease Classification. Electronics, 14(2), 305. https://doi.org/10.3390/electronics14020305
Vancouver
Tsai CM, Lee J. Dynamic Ensemble Learning with Gradient-Weighted Class Activation Mapping for Enhanced Gastrointestinal Disease Classification. Electronics. 2025;14(2):305. doi:10.3390/electronics14020305
BibTeX
@article{Tsai2025,
author = {Tsai, Chih Mao and Lee, Jiann-Der},
title = {Dynamic Ensemble Learning with Gradient-Weighted Class Activation Mapping for Enhanced Gastrointestinal Disease Classification},
journal = {Electronics},
year = {2025},
volume = {14},
number = {2},
pages = {305},
doi = {10.3390/electronics14020305}
}
EUVIMED