Image Translation-Based Unsupervised Cross-Modality Domain Adaptation for Medical Image Segmentation
This addresses the problem of domain shift in medical images for researchers and practitioners, enabling segmentation without target annotations, though it is incremental as it builds on existing image translation and self-training methods.
The paper tackles unsupervised cross-modality domain adaptation for medical image segmentation by using image translation to transform annotated source modality images into target modality images, achieving a mean Dice Similarity Coefficient of 0.8351 for vestibular schwannoma and 0.8098 for cochlea segmentation on a validation leaderboard.
Supervised deep learning usually faces more challenges in medical images than in natural images. Since annotations in medical images require the expertise of doctors and are more time-consuming and expensive. Thus, some researchers turn to unsupervised learning methods, which usually face inevitable performance drops. In addition, medical images may have been acquired at different medical centers with different scanners and under different image acquisition protocols, so the modalities of the medical images are often inconsistent. This modality difference (domain shift) also reduces the applicability of deep learning methods. In this regard, we propose an unsupervised crossmodality domain adaptation method based on image translation by transforming the source modality image with annotation into the unannotated target modality and using its annotation to achieve supervised learning of the target modality. In addition, the subtle differences between translated pseudo images and real images are overcome by self-training methods to further improve the task performance of deep learning. The proposed method showed mean Dice Similarity Coefficient (DSC) and Average Symmetric Surface Distance (ASSD) of $0.8351 \pm 0.1152$ and $1.6712 \pm 2.1948$ for vestibular schwannoma (VS), $0.8098 \pm 0.0233$ and $0.2317 \pm 0.1577$ for cochlea on the VS and cochlea segmentation task of the Cross-Modality Domain Adaptation (crossMoDA 2022) challenge validation phase leaderboard.