Cross-Domain Distribution Alignment for Segmentation of Private Unannotated 3D Medical Images
This addresses the challenge of data scarcity and privacy in medical imaging for segmentation tasks, though it is incremental as it builds on existing UDA methods.
The paper tackled the problem of training deep neural networks for 3D medical image segmentation without manual annotation due to privacy constraints, by introducing a source-free unsupervised domain adaptation method that generates pseudo-labels for self-training, achieving state-of-the-art performance on a real-world dataset.
Manual annotation of 3D medical images for segmentation tasks is tedious and time-consuming. Moreover, data privacy limits the applicability of crowd sourcing to perform data annotation in medical domains. As a result, training deep neural networks for medical image segmentation can be challenging. We introduce a new source-free Unsupervised Domain Adaptation (UDA) method to address this problem. Our idea is based on estimating the internally learned distribution of a relevant source domain by a base model and then generating pseudo-labels that are used for enhancing the model refinement through self-training. We demonstrate that our approach leads to SOTA performance on a real-world 3D medical dataset.