Radiomics extraction

[Carlos Martín]

Hello, I am perceiving differeneces betweet these two pipelines:

1.extracting radiomics from nifti files 

2.loading niftis to simpleITK images - setting resample pixel spacing - extract features. 

Additionally, if I create a dummy cubic mask of 3x3x3 pixels, shape features are correct.

Also if I save my processed sitk images as nifti with the original header and I extract directly from niftis, the extraction is working perfectly. I assume it can be something related to the affine matrix.

[j.v.gri...@nki.nl]

I'm not entirely sure what your process is in step 1, but it seems to me that you don't enable resampling there, while you do in step 2. This would explain the differences. If you can share the results (extracted features + the diagnostic features), I can check this.

Regards,

Joost

Verzonden vanaf mijn Samsung Galaxy-smartphone.

-------- Oorspronkelijk bericht --------

Van: Carlos Martín <carlos.mar...@gmail.com>

Datum: 22-01-20 09:16 (GMT+01:00)

Aan: pyradiomics <pyrad...@googlegroups.com>

Onderwerp: Radiomics extraction

--
You received this message because you are subscribed to the Google Groups "pyradiomics" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pyradiomics...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pyradiomics/CAMp5wgJ3Ve%2B6vocS1vuNQOSzhzNX_rp_ww6wPc9gyhMjV-tG6g%40mail.gmail.com.