Figure 2
From: Bladder Cancer Treatment Response Assessment in CT using Radiomics with Deep-Learning
Creating ROIs to train the DL-CNN. (a) ROIs were generated by combining regions from the pre- and post-treatment scan lesions. In this example, the pre-treatment stage was T3, and the post-treatment stage was T2. Therefore, the ROI was labeled as being greater than stage T0 after treatment. (b) ROI of a case that was stage T3 pre-treatment and stage T0 after treatment. (c) ROI of a case that was stage T2 pre-treatment and stage T4 post-treatment. Therefore the ROI was labeled as greater than stage T0 after treatment.
