We're published! Please check out the Technical Note here: https://www.sciencedirect.com/science/article/abs/pii/S1879850021000485 and reference this work if you find it useful

DOI:https://doi.org/10.1016/j.prro.2021.02.003

This code provides functionality for turning dicom images and RT structures into nifti files as well as turning prediction masks back into RT structures

Installation guide

pip install DicomRTTool

Highly recommend to go through the jupyter notebook in the Examples folder and to read the Wiki

Quick use guide

from DicomRTTool.ReaderWriter import DicomReaderWriter, ROIAssociationClass
Dicom_path = r'.some_path_to_dicom'
Dicom_reader = DicomReaderWriter(description='Examples', arg_max=True)
Dicom_reader.walk_through_folders(Dicom_path) # This will parse through all DICOM present in the folder and subfolders
all_rois = Dicom_reader.return_rois(print_rois=True) # Return a list of all rois present

Contour_names = ['tumor'] # Define what rois you want
associations = [ROIAssociationClass('tumor', ['tumor_mr', 'tumor_ct'])] # Any list of roi associations
Dicom_reader.set_contour_names_and_assocations(contour_names=Contour_names, associations=associations)

Dicom_reader.get_images_and_mask()

image_numpy = Dicom_reader.ArrayDicom
mask_numpy = Dicom_reader.mask
image_sitk_handle = Dicom_reader.dicom_handle
mask_sitk_handle = Dicom_reader.annotation_handle

Other interesting additions

Adding information to the Dicom_reader.series_instances_dictionary

from DicomRTTool.ReaderWriter import Tag
plan_pydicom_string_keys = {"MyNamedRTPlan": Tag((0x300a, 0x002))}
image_sitk_string_keys = {"MyPatientName": "0010|0010"}
Dicom_reader = DicomReaderWriter(description='Examples', arg_max=True, plan_pydicom_string_keys=plan_pydicom_string_keys, image_sitk_string_keys=image_sitk_string_keys)

If you find this code useful, please provide a reference to my github page for others www.github.com/brianmanderson , thank you!

Ring update allows for multiple rings to be represented correctly

Works on oblique images for masks and predictions*