Dicom Reader failing to interpret Z spacing.

[kgillette]

When using the import layer from series, Seg3D has had issues loading dicom data due to the interpretation of the z spacing. It seems to be artificially filling the z-direction of the image, maybe due to different formatting/information available in the header. I have tried with two different dicoms and had the same issue with both.

When I have loaded the dicoms into ITKsnap and outputted them as a nrrd, I could properly read the files.

[jessdtate]

can you provide an example dicom as a zip?

[kgillette]

The file is too big even when it is compressed. I have put the file in my SCI folder:

/home/sci/kgillette/test_dicom.zip

Unfortunately, this dicom is private and can only be used for testing purposes. It should not be used for any research or publications.

[jessdtate]

It looks like these data are CINE data. Is that what is loading incorrectly? Seg3D currently does not support multiple slices in the same position, as you would have with CINE. The dicoms should be split into different time steps before loading. I have written a python script that can do this, but it assumes all the files in a folder belong to the same CINE scan.

split_cine_dicom.py.zip

Is the spacing itself wrong or is it the time steps that is messed up?