Update BatchStructureSetConversion.py #141
Conversation
change the name of saved file to patient ID
| if not os.access(output_dir, os.F_OK): | ||
| os.mkdir(output_dir) | ||
| save_rtslices(output_dir, use_ref_image) |
There was a problem hiding this comment.
Please add the argument back, this would break the reference image feature.
There was a problem hiding this comment.
What's the difference between "save_rtslices(output_dir, use_ref_image)" and "save_rtslices(output_dir)". I know this is a previous issue, but after reading the post, I can't realize what is reference image.
There was a problem hiding this comment.
If you specify a reference image, then the saved NRRDs will have that dimension instead of the segments being cropped to their "effective extent". But even if you don't understand a feature, removing it is not a good idea :)
| slicer.mrmlScene.Clear(0) # clear the scene | ||
| DICOMUtils.loadPatientByUID(patient) | ||
| output_dir = os.path.join(output_folder,patient) | ||
| slicer.mrmlScene.Clear(0) # clear the scen |
| DICOMUtils.loadPatientByUID(patient) | ||
| output_dir = os.path.join(output_folder,patient) | ||
| slicer.mrmlScene.Clear(0) # clear the scen | ||
| studyList=db.studiesForPatient(patient) # code modified from https://www.slicer.org/wiki/Documentation/Nightly/ScriptRepository |
There was a problem hiding this comment.
This is more complicated than necessary. I think this would work better:
slicer.dicomDatabase.fieldForPatient('PatientID', patient)
In addition, I think the database UID should be kept, because in DICOM the patient ID is not necessarily unique, and it is unique together with the patient name, but it is fragile to use patient name in path. With the database patient UID though, it is unique within the Slicer database. Something like this:
patientID = slicer.dicomDatabase.fieldForPatient('PatientID', patient)
folderName = str(patient) + '_' + patientID
@lassoan What do you think?
There was a problem hiding this comment.
Patient ID can be really anything: it is often set to empty during image anonymization (even though it is not valid), can be too long to be used as filename, and may contain special characters that cannot be used in filenames. So, they are definitely much more risky to use than the database's patient UID.
I would recommend to make addint patient name+id to folder name optional and sanitize the string (remove special characters, limit maximum size, etc.).
change the name of saved file to patient ID