EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Explainable depth-wise and channel-wise fusion models for multi-class skin lesion classification

Humam AbuAlkebash iD, Radhwan A A Saleh iD, H Metin Ertunç

PLOS One · 22.01.2026 · 21 (1) · e0340901

DOI: 10.1371/journal.pone.0340901 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 22.01.2026. Verfasst von Humam AbuAlkebash und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Mechatronics Engineering, Palestine Polytechnic University, Hebron, Palestine. Hauptthemen laut MEDLINE: Deep Learning, Skin Diseases. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The clinical adoption of deep learning in dermatology requires models that are not only highly accurate but also transparent and trustworthy. To address this dual challenge, this study presents a systematic investigation into deep feature fusion, exploring how to effectively combine complementary representations from diverse neural network architectures. We design and rigorously evaluate six distinct fusion models, first investigating depth-wise and channel-wise strategies for integrating features from powerful Convolutional Neural Network (CNN) backbones, and subsequently incorporating the global contextual awareness of Vision Transformers (ViTs). Evaluated on the challenging 7-class HAM10000 dataset, our optimized architecture achieves a weighted average Precision, Recall, and F1 score of 90%, demonstrating superior diagnostic performance. Crucially, our comprehensive explainable AI (XAI) analysis using Grad-CAM and SHAP reveals that the fusion strategy directly dictates the model’s clinical interpretability; our most effective models learn to base their predictions on salient dermatological features, such as border irregularity and color variegation, in a manner that aligns with expert reasoning. This work provides a robust framework and valuable architectural insights for developing the next generation of high-performing, clinically reliable, and transparent AI-powered diagnostic tools.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningSkin DiseasesConvolutional Neural NetworksHumans

Autor:innen und Institutionen

  • Department of Mechatronics Engineering, Palestine Polytechnic University, Hebron, Palestine.
  • Faculty of Engineering and Information Technology, Taiz University, Taiz, Yemen.
  • Department of Software Engineering, Kocaeli University, Kocaeli, Türkiye.
  • Department of Mechatronics Engineering, Kocaeli University, Kocaeli, Türkiye.

ORCID-Profile: Humam AbuAlkebash, Radhwan A A Saleh

Zitieren

APA

AbuAlkebash, H., Saleh, R. A. A., & Ertunç, H. M. (2026). Explainable depth-wise and channel-wise fusion models for multi-class skin lesion classification. PLOS One, 21(1), e0340901. https://doi.org/10.1371/journal.pone.0340901

Vancouver

AbuAlkebash H, Saleh RAA, Ertunç HM. Explainable depth-wise and channel-wise fusion models for multi-class skin lesion classification. PLOS One. 2026;21(1):e0340901. doi:10.1371/journal.pone.0340901

BibTeX
@article{AbuAlkebash2026,
  author = {AbuAlkebash, Humam and Saleh, Radhwan A A and Ertunç, H Metin},
  title = {Explainable depth-wise and channel-wise fusion models for multi-class skin lesion classification},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {1},
  pages = {e0340901},
  doi = {10.1371/journal.pone.0340901}
}
Kostenlos starten →