Abstract:Introduction: Unsupervised domain adaptation (UDA) aims to adapt a model learned from the source domain to the target domain. Thus, the model can obtain transferable knowledge even in target domain that does not have ground truth in this way. In medical image segmentation scenarios, there exist diverse data distributions caused by intensity in homogeneities and shape variabilities. But multi source data may not be freely accessible, especially medical images with patient identity information. Methods: To tackle this issue, we propose a new multi-source and source-free (MSSF) application scenario and a novel domain adaptation framework where in the training stage, we only get access to the well-trained source domain segmentation models without source data. First, we propose a new dual consistency constraint which uses domain-intra and domain-inter consistency to filter those predictions agreed by each individual domain expert and all domain experts. It can serve as a high-quality pseudo label generation method and produce correct supervised signals for target domain supervised learning. Next, we design a progressive entropy loss minimization method to minimize the class-inter distance of features, which is beneficial to enhance domain-intra and domain-inter consistency in turn. Results: Extensive experiments are performed for retinal vessel segmentation under MSSF condition and our approach produces impressive performance. The sensitivity metric of our approach is highest and it surpasses other methods with a large margin. Discussion: It is the first attempt to conduct researches on the retinal vessel segmentation task under multi-source and source-free scenarios. In medical applications, such adaptation method can avoid the privacy issue. Furthermore, how to balance the high sensitivity and high accuracy need to be further considered.

Triple Consistency-based Self-ensembling Model for Unsupervised Domain Adaption in Medical Image Segmentation

Unsupervised Domain Adaptation for Medical Image Segmentation by Selective Entropy Constraints and Adaptive Semantic Alignment

Adversarial Consistency for Single Domain Generalization in Medical Image Segmentation

Transformation-Consistent Self-Ensembling Model for Semisupervised Medical Image Segmentation

Transformation Consistent Self-ensembling Model for Semi-supervised Medical Image Segmentation

Adaptive Entropy Regularization for Unsupervised Domain Adaptation in Medical Image Segmentation

A medical unsupervised domain adaptation framework based on Fourier transform image translation and multi-model ensemble self-training strategy

Style Consistency Unsupervised Domain Adaptation Medical Image Segmentation

Cycle and Self-Supervised Consistency Training for Adapting Semantic Segmentation of Aerial Images

Dual-consistency guidance semi-supervised medical image segmentation with low-level detail feature augmentation

Triple-task mutual consistency for semi-supervised 3D medical image segmentation

Dual consistent pseudo label generation for multi-source domain adaptation without source data for medical image segmentation

Semantic Consistent Unsupervised Domain Adaptation for Cross-Modality Medical Image Segmentation

Multi Task Consistency Guided Source-Free Test-Time Domain Adaptation Medical Image Segmentation

Dual-Task Self-supervision for Cross-modality Domain Adaptation

A multi-task unsupervised domain adaptation network for medical image segmentation

Unsupervised Cross-modality Cardiac Image Segmentation Via Disentangled Representation Learning and Consistency Regularization.

Tri-Directional Tasks Complementary Learning for Unsupervised Domain Adaptation of Cross-modality Medical Image Semantic Segmentation.

CyCMIS: Cycle-consistent Cross-domain Medical Image Segmentation Via Diverse Image Augmentation

Rethinking Adversarial Domain Adaptation: Orthogonal Decomposition for Unsupervised Domain Adaptation in Medical Image Segmentation

Towards collaborative appearance and semantic adaptation for medical image segmentation