AIM-Harvard / pyradiomics

Open-source python package for the extraction of Radiomics features from 2D and 3D images and binary masks. Support: https://discourse.slicer.org/c/community/radiomics
http://pyradiomics.readthedocs.io/
BSD 3-Clause "New" or "Revised" License
1.14k stars 492 forks source link

[FEAT EXTRACTION] Will discontinuous ROIs work? #750

Closed amirzolal closed 2 years ago

amirzolal commented 2 years ago

Hello, this is rather a general question regarding the functioning of the algorithm. I have also posted this on slicer discourse but there was no reaction over several days.

I have some ROIs that I use to extract image pixel intensity from images and would like to re-use these ROIs in feature detection. However, the ROIs are discontinuous (meaning that the a specific tissue is marked on different places in the scan, without there being a connection between those tissues).

The radiomics program (at least pyradiomics) does produce a csv when using this ROI.

Will the output be anyhow affected by the discontinuity, i mean does the ROI border affect the feature recognition somehow?

Thanks in advance for the answer,

Amir

giemmecci commented 2 years ago

I'm not part of the pyradiomics team, but the case discussed here seems similar to yours:

https://github.com/AIM-Harvard/pyradiomics/issues/689

JoostJM commented 2 years ago

@amirzolal, @giemmecci is correct.