Meta-Transfer Derm-Diagnosis: Exploring Few-Shot Learning and Transfer Learning for Skin Disease Classification in Long-Tail Distribution
IEEE Journal of Biomedical and Health Informatics, vol.30, no.4, pp.3132-3145, 2026 (SCI-Expanded, Scopus)
- Publication Type: Article / Article
- Volume: 30 Issue: 4
- Publication Date: 2026
- Doi Number: 10.1109/jbhi.2025.3615479
- Journal Name: IEEE Journal of Biomedical and Health Informatics
- Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, BIOSIS, Compendex, EMBASE, INSPEC, MEDLINE
- Page Numbers: pp.3132-3145
- Keywords: explainability, Few-shot learning, long-tail distribution, medical image classification, self supervised learning, skin disease classification, transfer learning, uncertainty estimation
- Hacettepe University Affiliated: Yes
Abstract
Building accurate models for rare skin diseases remains challenging due to the lack of sufficient labeled data and the inherently long-tailed distribution of available samples. These issues are further complicated by inconsistencies in how datasets are collected and their varying objectives. To address these challenges, we compare three learning strategies: episodic learning, supervised transfer learning, and contrastive self-supervised pretraining, within a few-shot learning framework. We evaluate five training setups on three benchmark datasets: ISIC2018, Derm7pt, and SD-198. Our findings show that traditional transfer learning approaches, particularly those based on MobileNetV2 and Vision Transformer (ViT) architectures, consistently outperform episodic and self-supervised methods as the number of training examples increases. When combined with batch-level data augmentation techniques such as MixUp, CutMix, and ResizeMix, these models achieve state-of-the-art performance on the SD-198 and Derm7pt datasets, and deliver highly competitive results on ISIC2018.