DICOM SEG converter assigns the same semantics to the converted segments

[fedorov]

You should at least alert the users of our library that every segment you save will be labeled as "Connective tissue".

pycad/pycad/converters/nifti_to_dicom_seg.py

Lines 68 to 77 in 2deae98

	description_segment_1 = hd.seg.SegmentDescription(
	segment_number=1,
	segment_label=segment_label,
	segmented_property_category=codes.cid7150.Tissue,
	segmented_property_type=codes.cid7166.ConnectiveTissue,
	algorithm_type=hd.seg.SegmentAlgorithmTypeValues.AUTOMATIC,
	algorithm_identification=algorithm_identification,
	tracking_uid=hd.UID(),
	tracking_id='test segmentation of computed tomography image'
	)

Some background reading about DICOM SEG: https://peerj.com/articles/2057/.

[amine0110]

Thank you for your comment!

This was a basic implementation of how to convert a NIFTI of one class to a DICOM SEG of one class. The implementation of multiclass was done, just haven't been published yet, as some optimizations are needed.

[fedorov]

The comment is not about multiclass - it is about assigning metadata, semantics of the segment. The converter must provide the interface for the user to specify the semantics of the segment being encoded.

[amine0110]

Just got your detailed explanation from LinkedIn, thank you for the details, I will take it into consideration.