Contrastive adapter training and consensus knowledge distillation for multi-source-free domain adaptation in skin...

Skin cancer diagnosis, particularly the differentiation of melanoma from benign nevi, is a vital yet challenging task due to the visual similarity between lesions. Although deep learning models such as convolutional neural networks (CNNs) and vision transformers (ViTs) have demonstrated promising performance, their effectiveness often deteriorates when applied to data from heterogeneous clinical sources. While conventional domain adaptation methods address domain shift, they require access to source data during adaptation, which is often infeasible due to privacy regulations. Multi-source-free unsupervised domain adaptation (MSFDA) addresses this limitation by leveraging multiple labeled source domains to generalize to an unlabeled target domain without requiring access to source data, making it suitable for privacy-sensitive medical settings. However, existing MSFDA methods rely on full backbone fine-tuning, leading to catastrophic forgetting and overfitting on small clinical datasets, and address domain shift at the aggregation stage without establishing a shared domain-invariant feature space. Furthermore, their reliance on hard pseudo-labels or confidence-weighted aggregation introduces noisy supervision signals under domain shift. To address these limitations, we propose CAT-CKD, consisting of two components: (1) contrastive adapter training (CAT), which trains lightweight ConvPass adapters within a frozen ViT backbone using supervised contrastive learning (SCL) to establish a shared domain-invariant feature space before source-specific model training, and (2) consensus knowledge distillation (CKD), which aggregates logits from multiple source models into a consensus supervisory signal and adapts a student model on unlabeled target data using KL divergence. Experiments on five publicly available skin lesion datasets show that CAT-CKD achieves an average AUROC of 86.1%, outperforming existing MSFDA methods while requiring only 4.3M trainable parameters. The code for this paper is available at https://github.com/A-Abedi/CAT_CKD.