UPL-SFDA: Uncertainty-aware Pseudo Label Guided Source-Free Domain Adaptation for Medical Image Segmentation

TL;DR

This paper introduces UPL-SFDA, a novel uncertainty-aware pseudo label guided source-free domain adaptation method that significantly improves medical image segmentation across various datasets by leveraging pseudo labels, model perturbations, and entropy minimization.

Contribution

The paper proposes a new SFDA approach with Target Domain Growing, Twice Forward pass Supervision, and entropy regularization, enhancing pseudo label reliability and segmentation accuracy.

Findings

Improved Dice scores by over 5 percentage points on multiple datasets.

Outperformed existing state-of-the-art SFDA methods.

Validated across diverse medical imaging tasks.

Abstract

Domain Adaptation (DA) is important for deep learning-based medical image segmentation models to deal with testing images from a new target domain. As the source-domain data are usually unavailable when a trained model is deployed at a new center, Source-Free Domain Adaptation (SFDA) is appealing for data and annotation-efficient adaptation to the target domain. However, existing SFDA methods have a limited performance due to lack of sufficient supervision with source-domain images unavailable and target-domain images unlabeled. We propose a novel Uncertainty-aware Pseudo Label guided (UPL) SFDA method for medical image segmentation. Specifically, we propose Target Domain Growing (TDG) to enhance the diversity of predictions in the target domain by duplicating the pre-trained model's prediction head multiple times with perturbations. The different predictions in these duplicated heads are used to obtain pseudo labels for unlabeled target-domain images and their uncertainty to identify reliable pseudo labels. We also propose a Twice Forward pass Supervision (TFS) strategy that uses reliable pseudo labels obtained in one forward pass to supervise predictions in the next forward pass. The adaptation is further regularized by a mean prediction-based entropy minimization term that encourages confident and consistent results in different prediction heads. UPL-SFDA was validated with a multi-site heart MRI segmentation dataset, a cross-modality fetal brain segmentation dataset, and a 3D fetal tissue segmentation dataset. It improved the average Dice by 5.54, 5.01 and 6.89 percentage points for the three tasks compared with the baseline, respectively, and outperformed several state-of-the-art SFDA methods.