EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Calibration-Aware Heterogeneous Multimodal Ensemble for Multi-Class Skin Lesion Diagnosis

Ozan Alptekin iD, Feristah Dalkilic iD, Zerrin Isik iD

Diagnostics · 01.11.2026 · 16 (21) · 3613

DOI: 10.3390/diagnostics16213613 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Diagnostics, online erschienen am 01.11.2026. Verfasst von Ozan Alptekin und 2 weiteren Autor:innen. Beteiligte Einrichtung: Graduate School of Natural and Applied Sciences, Dokuz Eylul University, 35390 İzmir, Türkiye. Die Arbeit nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Reliable classification of skin lesions is essential for early melanoma detection; however, substantial visual overlap among lesion types and pronounced class imbalance continue to limit the performance and reliability of existing automated systems. These challenges motivate the integration of complementary visual representations with clinical metadata to improve discrimination across lesion classes.

Methods

This study introduces a calibrated multimodal ensemble learning framework for skin lesion classification. The proposed framework combines visual features extracted from Swin Transformer, EVA-02, and CoAtNet architectures with patient age and anatomical region information. A two-stage adaptive temperature scaling is applied to calibrate the predicted probabilities of the ensemble.

Results

Experimental evaluation on the International Skin Imaging Collaboration (ISIC 2019) dataset demonstrates strong classification performance, achieving a macro-F1 score of 0.91 and a macro-AUC of 0.99. The framework further showed robust generalization on the official ISIC 2019 benchmark set, attaining a balanced accuracy of 0.65. The external validation on the PH2 dataset yielded a macro-F1 score of 0.93 and an ROC-AUC of 0.96.

Conclusions

These findings demonstrate that combining heterogeneous visual representations, clinical metadata, and calibration-aware ensemble learning provides a robust and trustworthy solution for skin lesion classification and computer-aided dermatological diagnosis.

Schlagwörter

Stichwörter der Autor:innen

hybrid deep learningmelanoma diagnosismultimodal fusionprobability calibrationskin lesion classificationtransformer models

Autor:innen und Institutionen

  • Graduate School of Natural and Applied Sciences, Dokuz Eylul University, 35390 İzmir, Türkiye.
  • Department of Computer Engineering, Dokuz Eylul University, 35390 İzmir, Türkiye.
  • Bioinformatics Group, Department of Computer Science, Vrije Universiteit Amsterdam, 1081 HV Amsterdam, The Netherlands.

ORCID-Profile: Ozan Alptekin, Feristah Dalkilic, Zerrin Isik

Zitieren

APA

Alptekin, O., Dalkilic, F., & Isik, Z. (2026). Calibration-Aware Heterogeneous Multimodal Ensemble for Multi-Class Skin Lesion Diagnosis. Diagnostics, 16(21), 3613. https://doi.org/10.3390/diagnostics16213613

Vancouver

Alptekin O, Dalkilic F, Isik Z. Calibration-Aware Heterogeneous Multimodal Ensemble for Multi-Class Skin Lesion Diagnosis. Diagnostics. 2026;16(21):3613. doi:10.3390/diagnostics16213613

BibTeX
@article{Alptekin2026,
  author = {Alptekin, Ozan and Dalkilic, Feristah and Isik, Zerrin},
  title = {Calibration-Aware Heterogeneous Multimodal Ensemble for Multi-Class Skin Lesion Diagnosis},
  journal = {Diagnostics},
  year = {2026},
  volume = {16},
  number = {21},
  pages = {3613},
  doi = {10.3390/diagnostics16213613}
}
Kostenlos starten →