An Ingeniously Designed Skin Lesion Classification Model Across Clinical and Dermatoscopic Datasets
Diagnostics · 11.08.2025 · 15 (16) · 2011
DOI: 10.3390/diagnostics15162011 ↗Kurz gefasst
Fachartikel in der Zeitschrift Diagnostics, online erschienen am 11.08.2025. Verfasst von Ying Huang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Dermatovenereology, West China Hospital, Sichuan University, Chengdu 610041, China. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: 135 project for disciplines of excellence-Clinical Research Incubation Project, National Key Research and Development Program of China, 135 projects for disciplines of excellence—Clinical Research Incubation Project, West China Hospital, Sichuan University.
Abstract (Originalsprache)
Background
Skin cancer diagnosis faces critical challenges due to the visual similarity of lesions and dataset limitations.
Methods
This study introduces HybridSkinFormer, a robust deep learning model designed to classify skin lesions from both clinical and dermatoscopic images. The model employs a two-stage architecture: a multi-layer ConvNet for local feature extraction and a residual-learnable multi-head attention module for global context fusion. A novel activation function (StarPRelu) and Enhanced Focal Loss (EFLoss) address neuron death and class imbalance, respectively.
Results
Evaluated on a hybrid dataset (37,483 images across nine classes), HybridSkinFormer achieved state-of-the-art performance with an overall accuracy of 94.2%, a macro precision of 91.1%, and a macro recall of 91.0%, outperforming nine CNN and ViT baselines.
Conclusions
Its ability to handle multi-modality data and mitigate imbalance highlights its clinical utility for early cancer detection in resource-constrained settings.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Dermatovenereology, West China Hospital, Sichuan University, Chengdu 610041, China.
- School of Information and Software Engineering, University of Electronic Science and Technology of China (UESTC), Chengdu 610054, China.
ORCID-Profile: Zhishuo Zhang, Xin Ran, Kaiwen Zhuang, Yuping Ran
Förderung
- 135 project for disciplines of excellence-Clinical Research Incubation Project
- National Key Research and Development Program of China
- 135 projects for disciplines of excellence—Clinical Research Incubation Project, West China Hospital, Sichuan University
Zitieren
APA
Huang, Y., Zhang, Z., Ran, X., Zhuang, K., & Ran, Y. (2025). An Ingeniously Designed Skin Lesion Classification Model Across Clinical and Dermatoscopic Datasets. Diagnostics, 15(16), 2011. https://doi.org/10.3390/diagnostics15162011
Vancouver
Huang Y, Zhang Z, Ran X, Zhuang K, Ran Y. An Ingeniously Designed Skin Lesion Classification Model Across Clinical and Dermatoscopic Datasets. Diagnostics. 2025;15(16):2011. doi:10.3390/diagnostics15162011
BibTeX
@article{Huang2025,
author = {Huang, Ying and Zhang, Zhishuo and Ran, Xin and Zhuang, Kaiwen and Ran, Yuping},
title = {An Ingeniously Designed Skin Lesion Classification Model Across Clinical and Dermatoscopic Datasets},
journal = {Diagnostics},
year = {2025},
volume = {15},
number = {16},
pages = {2011},
doi = {10.3390/diagnostics15162011}
}
EUVIMED