Source-Free Domain Adaptive Fundus Image Segmentation with Denoised Pseudo-Labeling
This addresses privacy concerns in medical imaging by enabling domain adaptation without accessing source data, though it is incremental as it builds on existing pseudo-labeling techniques.
The paper tackles source-free unsupervised domain adaptation for fundus image segmentation, where only a pre-trained source model and unlabeled target data are available, by introducing a denoised pseudo-labeling method with pixel-level and class-level denoising schemes, achieving comparable or higher performance than state-of-the-art source-dependent methods.
Domain adaptation typically requires to access source domain data to utilize their distribution information for domain alignment with the target data. However, in many real-world scenarios, the source data may not be accessible during the model adaptation in the target domain due to privacy issue. This paper studies the practical yet challenging source-free unsupervised domain adaptation problem, in which only an existing source model and the unlabeled target data are available for model adaptation. We present a novel denoised pseudo-labeling method for this problem, which effectively makes use of the source model and unlabeled target data to promote model self-adaptation from pseudo labels. Importantly, considering that the pseudo labels generated from source model are inevitably noisy due to domain shift, we further introduce two complementary pixel-level and class-level denoising schemes with uncertainty estimation and prototype estimation to reduce noisy pseudo labels and select reliable ones to enhance the pseudo-labeling efficacy. Experimental results on cross-domain fundus image segmentation show that without using any source images or altering source training, our approach achieves comparable or even higher performance than state-of-the-art source-dependent unsupervised domain adaptation methods.