Contrastive meta-domain adaptation for robust skin lesion classification across clinical and acquisition conditions

2026-02-23

AI Summary

This work addresses the significant performance degradation of skin lesion classification models when deployed across diverse clinical settings and imaging conditions, primarily due to domain shift and visual artifacts. To mitigate this issue, the authors propose a contrastive transfer learning approach grounded in the concept of visual meta-domains. By integrating meta-domain modeling with contrastive learning, the method enables robust representation transfer from large dermoscopic datasets to real-world clinical image domains. The framework effectively leverages multi-source dermatological imaging data, substantially improving classification accuracy across multiple clinical datasets. This advancement narrows the performance gap between dermoscopic and clinical images and enhances the model’s cross-domain generalization capability.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Multi-modal Vision
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Deep learning models for dermatological image analysis remain sensitive to acquisition variability and domain-specific visual characteristics, leading to performance degradation when deployed in clinical settings. We investigate how visual artifacts and domain shifts affect deep learning-based skin lesion classification. We propose an adaptation strategy, grounded in the idea of visual meta-domains, that transfers visual representations from larger dermoscopic datasets into clinical image domains, thereby improving generalization robustness. Experiments across multiple dermatology datasets show consistent gains in classification performance and reduced gaps between dermoscopic and clinical images. These results emphasize the importance of domain-aware training for deployable systems.

Problem

• domain shift
• skin lesion classification
• acquisition variability
• clinical deployment
• visual artifacts

Innovation

• contrastive learning
• meta-domain adaptation
• domain generalization
• skin lesion classification
• visual artifacts