Jakob Wasserthal
Jakob Wasserthal
The running of the segmentation model for the contrast phase segmentation is quite fast since it only uses a few ROIs and a low resolution model. The runtime you save...
The coronary_arteries model does not work as robustly as the other totalsegmentator models as it was trained only on a small dataset. It only works well if you have a...
TotalSegmentator expects as input the original CT image and typically works well with any image setting. So you do not need anything specifically. You have to make sure that you...
I would recommend not using the --preprocess flag at all. Instead do the registration to MNI space before yourself. Then only use those images in MNI space for TractSeg. Then...
The code is having problems downloading the model weights from zenodo. Sometimes zenodo is having download issues. I would suggest to try it again another time and/or from another internet...
Is this the complete error message? There seems to be missing something.
For the previous commands (output type tract_segmentation and endings_segmentations) you have to specify the same output folder. And for output_type TOM as well. The correct subfolders are automatically generated. Just...
Unfortunately supporting this would require a large refactoring of the code base. Therefore, this did not happen so far.
Thanks for posting this error. I will get a M1 Mac Book soon and then I hopefully can reproduce this error and try to solve it. Now I do not...
Is there any update if numpy > 2 is getting supported?