An Explainable Vision Transformer Framework for Skin Lesion Classification Using Dual-Map Fusion Strategy
International Journal of Computer Applications · 20.08.2026 · 187 (135) · 35-42
DOI: 10.5120/ijcac919f6ae2058 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Computer Applications, online erschienen am 20.08.2026. Verfasst von Tu Thanh Tri und 3 weiteren Autor:innen.
Abstract (Originalsprache)
Accurate and interpretable skin lesion classification is essential for early melanoma diagnosis and clinical decision support.This study proposes an explainable computer-aided diagnosis framework based on a Vision Transformer (ViT) for binary classification of benign and malignant skin lesions.To improve model transparency, Grad-CAM and transformer attention maps are integrated through a Dual-Map Fusion strategy, providing complementary local and global visual explanations of the model's predictions.The proposed framework was evaluated using a combined HAM10000 and ISIC dermoscopic image dataset.Experimental results demonstrated an overall classification accuracy of 92.55% and a malignant lesion recall of 94.68%, indicating reliable diagnostic performance with a reduced risk of missed malignant cases.In addition, the fused explanation maps provided more informative and interpretable visual evidence than Grad-CAM or attention maps alone, facilitating a better understanding of the model's decisionmaking process.The proposed framework combines the strong classification capability of Vision Transformer with enhanced explainability, providing an effective and transparent decisionsupport tool for automated skin lesion diagnosis.These findings demonstrate the potential of explainable transformerbased models for reliable clinical application in dermatological image analysis.
Zitieren
APA
Tri, T. T., Nga, D. T. T., Toan, C. P., & Lien, D. K. (2026). An Explainable Vision Transformer Framework for Skin Lesion Classification Using Dual-Map Fusion Strategy. International Journal of Computer Applications, 187(135), 35-42. https://doi.org/10.5120/ijcac919f6ae2058
Vancouver
Tri TT, Nga DTT, Toan CP, Lien DK. An Explainable Vision Transformer Framework for Skin Lesion Classification Using Dual-Map Fusion Strategy. International Journal of Computer Applications. 2026;187(135):35-42. doi:10.5120/ijcac919f6ae2058
BibTeX
@article{Tri2026,
author = {Tri, Tu Thanh and Nga, Duong Thi Thuy and Toan, Chau Phuong and Lien, Dang Kim},
title = {An Explainable Vision Transformer Framework for Skin Lesion Classification Using Dual-Map Fusion Strategy},
journal = {International Journal of Computer Applications},
year = {2026},
volume = {187},
number = {135},
pages = {35--42},
doi = {10.5120/ijcac919f6ae2058}
}
EUVIMED