In Slicer, I have segment done on CT images, but registration for CT and MRI is not good (these two are too far away). So I manually transform MRI to align with CT and harden the transform. Then, generate labelmap from the segment, and try to extract features using Radiomics, transformed MRI as input image, labelmap as input region. Not sure this is possible/a proper way in Radiomics, I can get some numbers e.g. min, max, mean… using “Label statistics”, but Radiomics keeps giving me errors:
ValueError: Bounding box of ROI is larger than image space:
ROI bounds (x, y, z image coordinate space) ((146.8418475884187, 83.86141228352409, 7.687500085149379), (227.50923956172693, 159.2217389954305, 22.27083297696697))
Image Size (352, 352, 22)
I think it might because some small part of segment is out of boundary of MRI (the red circle part)? Any suggestion to make this work without change the segmentation?