how to read non uniform sampling dicom file

[JohnMasoner]

I read the non-uniformly sampled dcm files as a list and then concateated them. There will be some deviation when I read RT Structure using the same method. This seems to be caused by the difference in spacing. The code of my method is as follows,

# this is a list cover different spacing SITKImage
patient_img = sitk_get_image(patient_dir, ignore_nonuniform=False)
new = []
new_img = []
for img in  patient_img:
    # this is RT Structure.
    patient_ste = rtstruct2stencil(img, patient_rts, roi_names).astype(np.uint8)[..., 0]
    new.append(patient_ste)
    new_img.append(sitk.GetArrayFromImage(img))
patient_ste = np.concatenate(new, axis=0)
patient_img = np.concatenate(new_img, axis=0)

I just use SimpleITK in the above method, MONAI is there a better way to read non uniform sampling dicom and RT Structure?

or, my model using NifTI file format(nii.gz) as the input data format , because the DCM file format needs some time to reading it, Is there any way to shorten the time to read the data and make sure the data is correct?

[wyli]

thanks for reporting, dicom rt is not supported at the moment, we may host a separate dicom utility codebase for this, but this is still under discussion.

[JohnMasoner]

thanks for your reply. Looking forward to this codebase.