# Validity of Radiomics Fractional Bin Width

**URL:** <https://discourse.slicer.org/t/validity-of-radiomics-fractional-bin-width/13877>\
**Category:** Radiomics\
**Tags:** mri, radiomics, pyradiomics\
**Created:** [October 6, 2020, 2:03am UTC](https://discourse.slicer.org/t/validity-of-radiomics-fractional-bin-width/13877 "2020-10-06T02:03:54Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![furMan](https://avatars.discourse-cdn.com/v4/letter/f/ea666f/32.png) [@furMan](https://discourse.slicer.org/u/furMan)\
**Post date:** [October 6, 2020, 2:03am UTC](https://discourse.slicer.org/t/validity-of-radiomics-fractional-bin-width/13877/1 "2020-10-06T02:03:54Z")

</div>

Operating system: Windows 10  
Slicer version:4.10.2  
Expected behavior:  
Actual behavior:

I’m analyzing MR images, and am normalizing the images using the following method:

1. Subtract the mean of the non-zero intensities
2. Divide by the standard deviation of the non-zero intensities

This results in a smaller range of intensities (between 0 and on order of ~10). In order to have a reasonable number of bins, I’ve been using fractional bin width values; e.g., 0.25, for the Radiomics input.

I see that fractional values are accepted, but wanted to verify that using fractional bin widths is valid. Thank you for any advice.

---

<div class="post-metadata">

**Author:** ![fedorov](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/fedorov/32/14_2.png) [@fedorov](https://discourse.slicer.org/u/fedorov)\
**Post date:** [October 6, 2020, 2:06am UTC](https://discourse.slicer.org/t/validity-of-radiomics-fractional-bin-width/13877/2 "2020-10-06T02:06:08Z")

</div>

Yes, fractional bin width is valid.