Adipose Tissue Segmentation in Unlabeled Abdomen MRI using Cross Modality Domain Adaptation
This work addresses the need for automated fat quantification in MRI to avoid radiation exposure from CT, though it is incremental as it adapts existing GAN techniques to a specific medical imaging task.
The authors tackled the problem of labor-intensive adipose tissue segmentation in abdomen MRI by proposing a cross-modality domain adaptation method using a cycle GAN to generate synthetic CT images, achieving average success scores of 3.80/5 for visceral and 4.54/5 for subcutaneous fat segmentation as evaluated by radiologists.
Abdominal fat quantification is critical since multiple vital organs are located within this region. Although computed tomography (CT) is a highly sensitive modality to segment body fat, it involves ionizing radiations which makes magnetic resonance imaging (MRI) a preferable alternative for this purpose. Additionally, the superior soft tissue contrast in MRI could lead to more accurate results. Yet, it is highly labor intensive to segment fat in MRI scans. In this study, we propose an algorithm based on deep learning technique(s) to automatically quantify fat tissue from MR images through a cross modality adaptation. Our method does not require supervised labeling of MR scans, instead, we utilize a cycle generative adversarial network (C-GAN) to construct a pipeline that transforms the existing MR scans into their equivalent synthetic CT (s-CT) images where fat segmentation is relatively easier due to the descriptive nature of HU (hounsfield unit) in CT images. The fat segmentation results for MRI scans were evaluated by expert radiologist. Qualitative evaluation of our segmentation results shows average success score of 3.80/5 and 4.54/5 for visceral and subcutaneous fat segmentation in MR images.