AUTHOR=Cai Binke , Ma Liyan , Sun Yan 

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

JOURNAL=Frontiers in Neuroscience

VOLUME=Volume 17 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2023.1209132

DOI=10.3389/fnins.2023.1209132

ISSN=1662-453X

ABSTRACT=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 inhomogeneities and shape variabilities. But the
multi source data may not be freely accessible, especially these medical images with patient
identity information. To tackle this issue, we propose a new Multi Source and Source Free (MSSF)
application scenario and a novel domain adaptation framework where in training stage we only
get access to the well-trained source domain segmentation models without source data. First,
we proposed 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. Extensive experiments are performed
for retinal vessel segmentation under Multi Source and Source Free (MSSF) condition and our
approach produce impressive performance.