Meta-Transfer Derm-Diagnosis: Exploring Few-Shot Learning and Transfer Learning for Skin Disease Classification in Long-Tail Distribution

TL;DR

This paper investigates the effectiveness of transfer learning, few-shot learning, and contrastive self-supervised pretraining for classifying rare skin diseases, demonstrating that transfer learning with data augmentation yields state-of-the-art results.

Contribution

It compares three learning strategies within a few-shot framework and shows that transfer learning with data augmentation outperforms other methods on skin disease datasets.

Findings

Transfer learning with MobileNetV2 and ViT outperforms episodic and self-supervised methods.

Data augmentation techniques like MixUp, CutMix, and ResizeMix improve model performance.

Achieved state-of-the-art results on SD-198 and Derm7pt datasets.

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. All the source codes related to this work will be made publicly available soon at the provided URL.