Domain Adaptation for the Segmentation of Confidential Medical Images
This work is significant for medical imaging practitioners as it offers a solution for adapting segmentation models to new modalities without needing to re-annotate data or access sensitive source patient data, which is crucial given data sharing restrictions.
This paper addresses the challenge of domain shift in medical image segmentation when source and target datasets originate from different modalities. The authors propose an unsupervised domain adaptation (UDA) algorithm that does not require access to source domain data during adaptation, achieving competitive performance against state-of-the-art methods on two medical image segmentation datasets.
Convolutional neural networks (CNNs) have led to significant improvements in the semantic segmentation of images. When source and target datasets come from different modalities, CNN performance suffers due to domain shift. In such cases data annotation in the target domain becomes necessary to maintain model performance. To circumvent the re-annotation process, unsupervised domain adaptation (UDA) is proposed to adapt a model to new modalities using solely unlabeled target data. Common UDA algorithms require access to source domain data during adaptation, which may not be feasible in medical imaging due to data sharing restrictions. In this work, we develop an algorithm for UDA where the source domain data is inaccessible during target adaptation. Our approach is based on encoding the source domain information into an internal distribution that is used to guide adaptation in the absence of source samples. We demonstrate the effectiveness of our algorithm by comparing it to state-of-the-art medical image semantic segmentation approaches on two medical image semantic segmentation datasets.