Self-supervised approach for organs at risk segmentation of abdominal CT images

Accurate segmentation of organs at risk is essential for radiation therapy planning. However, manual time-consuming and prone to inter intra-observer variability. This study proposes a self-supervision based attention UNet model OAR abdominal CT images. The utilizes mechanism train itself without the need annotations. used highlight important features suppress irrelevant ones, thus improving model’s accuracy. evaluated on dataset 100 scans compared its perfor mance with state-of-the-art methods. Our results show that proposed got comparable performance in terms dice similarity coefficient. More over, inference time much faster than traditional methods, making it promising tool clinical use.