CycleSeg: MR-to-CT Synthesis and Segmentation Network for Prostate Radiotherapy Treatment Planning
Huan Minh Luu1, Gyu-sang Yoo2, Dong-Hyun Kim1, Won Park2, and Sung-Hong Park1
1Magnetic Resonance Imaging Laboratory, Department of Bio and Brain Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Korea, Republic of, 2Department of Radiation Oncology, Samsung Medical Center, Seoul, Korea, Republic of
In this study, we proposed CycleSeg, a CycleGan-based network to produce synthetic CT images and organ segmentation for prostate cancer radiotherapy planning. Experiments showed that the network can generate realistic synthetic CT images and accurate segmentation.
Figure 1: Architecture
of CycleSeg,
which is a combination of CycleGAN and 2 Unets. The notations in this figure
correspond to those in Equation (1) in the text.
Figure 4: Qualitative comparison of organ
segmentation for a representative slice from the four networks (detail in
text). The segmentation is displayed as contour.