Comparative Analysis of Transfer Learning Models for Multi-Class Skin Disease Classification on HAM10000

DOI

10.2991/978-94-6239-664-7_97How to use a DOI?

Keywords

Transfer learning; Fine-tuning; Skin disease; InceptionResNetV2; NASNetLarge; EfficientNetV2L; DenseNet201; ResNet101V2; VGG16

Abstract

Skin diseases are a major public health issue worldwide, requiring early diagnosis and intervention to prevent serious consequences, including melanoma-related deaths. However, the traditional dermatoscopic diagnosis of skin diseases is often time-consuming, subjective, and depends on the availability of skilled dermatologists. To improve the diagnosis of skin diseases, the current research provides a comprehensive investigation of the latest transfer learning-based convolutional neural network (CNN) models for multi-class skin disease classification using dermatoscopic images. Six pre-trained deep learning models, including InceptionResNetV2, NASNetLarge, EfficientNetV2L, DenseNet201, ResNet101V2, and VGG16, were fine-tuned and tested on the HAM10000 dataset, consisting of 10,015 images divided into seven classes of skin disease. A series of experiments were conducted to evaluate the effectiveness of the proposed models in multi-class skin disease classification. Experimental results show that fine-tuning significantly improves model convergence and generalization. Of all the models tested, EfficientNetV2L achieved the highest classification accuracy, test accuracy of 87.92%, lowest loss of 4.76%, and highest precision, recall, and F1-score. The experimental results support the fact that the latest CNN models, which are optimized for scaling and feature extraction, are better than traditional models for multi-class skin disease classification. This research provides valuable insights for choosing effective deep learning models for multi-class skin disease classification, and the results show the effectiveness of EfficientNet-based models for developing reliable multi-class skin disease classification models.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Partha Jyoti Roy
AU  - Faisal Islam
PY  - 2026
DA  - 2026/06/08
TI  - Comparative Analysis of Transfer Learning Models for Multi-Class Skin Disease Classification on HAM10000
BT  - Proceedings of the International Conference on Intelligent Data Analysis and Applications (IDAA 2025)
PB  - Atlantis Press
SP  - 1437
EP  - 1452
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-664-7_97
DO  - 10.2991/978-94-6239-664-7_97
ID  - Roy2026
ER  -