EUVIMEDDécouvrir EUVIMED
Open access · CC BY 4.0

Ensemble Artificial Intelligence for Dermoscopic Decision Support: Development and External Validation of the HAM20000 Dataset

Ming-Hseng Tseng, Wen-Da Liu, Yu-Hsien Chen

Electronics · 2026-05-31 · 15 (11) · 2367

DOI: 10.3390/electronics15112367 ↗

At a glance

Journal article in Electronics, published online on 2026-05-31. Written by Ming-Hseng Tseng and 2 co-authors. 32 references. The full text is freely available (open access, CC BY 4.0). Funded by: National Science and Technology Council, Taiwan, R.O.C..

RésuméRéférences bibliographiquesCiteRelated studies

Abstract (original language)

Artificial intelligence (AI) has shown strong potential in dermoscopic image classification; however, reliable clinical deployment remains challenging due to limited generalizability across populations and institutions. This challenge is further compounded by the performance saturation of single models: continued architectural scaling often results in diminishing returns within the training domain and fails to yield corresponding gains in cross-domain robustness. Previous work often relies on a single benchmark dataset and evaluate models under matched data distributions, which can overestimate real-world performance. In this study, we examine the robustness and generalizability of dermoscopic AI systems under realistic deployment conditions. We curate HAM20000, a controlled expansion of the HAM10000 dataset integrating multiple public ISIC releases from 2017 to 2024, designed as a stress-test dataset to analyze performance saturation and architectural sensitivity. A total of 27 pretrained deep learning models, including convolutional neural networks, Vision Transformers, and hybrid CNN–Transformer architectures, were systematically evaluated. To mitigate the limitations of single-model optimization, we applied greedy ensemble selection to construct a compact heterogeneous ensemble using soft voting. All models were trained and selected exclusively on HAM20000 and evaluated under strict zero-shot external validation on two independent datasets: CSMUH (East Asian population) and BCN20000 (European cohort). Although the proposed ensemble achieved over 93% accuracy on the HAM20000 dataset, its performance declined substantially under external validation, highlighting a pronounced cross-population domain gap. On the BCN20000 external validation set, the ensemble reached 62% accuracy, outperforming the strongest individual model (59%) by 3%. On the CSMUH external validation set, the proposed GES achieved an accuracy of approximately 57%, demonstrating improved performance over representative single-model baselines, although it did not exceed the best-performing individual architecture in this dataset. These results indicate that, despite a marked absolute performance drop under distribution shift, the ensemble provides a modest yet consistent improvement over single-model baselines rather than a substantial performance leap. This finding underscores the importance of rigorous external validation and suggests that heterogeneous ensemble learning represents a practical and promising strategy for achieving incremental robustness gains in clinically applicable dermoscopic decision support systems.

Funding

  • National Science and Technology Council, Taiwan, R.O.C.

Cite

APA

Tseng, M., Liu, W., & Chen, Y. (2026). Ensemble Artificial Intelligence for Dermoscopic Decision Support: Development and External Validation of the HAM20000 Dataset. Electronics, 15(11), 2367. https://doi.org/10.3390/electronics15112367

Vancouver

Tseng M, Liu W, Chen Y. Ensemble Artificial Intelligence for Dermoscopic Decision Support: Development and External Validation of the HAM20000 Dataset. Electronics. 2026;15(11):2367. doi:10.3390/electronics15112367

BibTeX
@article{Tseng2026,
  author = {Tseng, Ming-Hseng and Liu, Wen-Da and Chen, Yu-Hsien},
  title = {Ensemble Artificial Intelligence for Dermoscopic Decision Support: Development and External Validation of the HAM20000 Dataset},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {11},
  pages = {2367},
  doi = {10.3390/electronics15112367}
}
Start for free →