Fig. 6From: Pelvic U-Net: multi-label semantic segmentation of pelvic organs at risk for radiation therapy anal cancer patients using a deeply supervised shuffle attention convolutional neural networkObservational evaluation results of pelvic organs at risk (OAR) segmentations. Segmentation predictions for \({\text{n}} = 15\) test patients were inspected by three clinical experts and rated on a scale between 1–4 (Excellent—Not acceptable). Results are presented as the mean ratings for each structure for both the uncorrected segmentations (stage 1) and the manually corrected segmentations (stage 2)Back to article page