# DWI radiomic features looking awkward?
**URL:** https://discourse.slicer.org/t/dwi-radiomic-features-looking-awkward/41782
**Category:** Support
**Tags:** segmentation, radiomics, pyradiomics
**Created:** [February 19, 2025, 9:21pm UTC](https://discourse.slicer.org/t/dwi-radiomic-features-looking-awkward/41782 "2025-02-19T21:21:15Z")
**Posts on this page:** 2
**Page:** 1
### Author:  [@Fazilhan\_Altintas](https://discourse.slicer.org/u/Fazilhan_Altintas)
#### Post date: [February 19, 2025, 9:21pm UTC](https://discourse.slicer.org/t/dwi-radiomic-features-looking-awkward/41782/1 "2025-02-19T21:21:15Z")
Hello Dear 3d Slicer Community,
I was segmenting a lesion on mulitple mri sequances to extract radiomic features. After I did I realized something very weird. For example I drew a mask on DWI and then slapped it onto adc. When I extracted radiomic features from both, meshvolume, which should be similar for both, was extremely different. This never happens with t2 t1 adc or contrast segmentations just with DWI. And this problem is prevelant across all other DWI segmentations and radiomic extractions I have produced (not just for meshvolume, though it was what made me suspect in the first place). Is there a secret sauce to using radiomics module on diffusion weighted imaging? What am I doing wrong. I greatly appreciate any help
---
### Author:  [@pieper](https://discourse.slicer.org/u/pieper)
#### Post date: [February 19, 2025, 10:12pm UTC](https://discourse.slicer.org/t/dwi-radiomic-features-looking-awkward/41782/2 "2025-02-19T22:12:02Z")
This sounds like something you should try to work out at the pyradiomics level since SlicerRadiomics is just a wrapper.
> **[AIM-Harvard/pyradiomics](https://github.com/AIM-Harvard/pyradiomics/issues)**
>
> 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 - AIM-Harvard/pyradiomics