# Problem with GLCM features

**URL:** https://discourse.slicer.org/t/problem-with-glcm-features/39499
**Category:** Radiomics
**Tags:** python
**Created:** [October 2, 2024, 5:26pm UTC](https://discourse.slicer.org/t/problem-with-glcm-features/39499 "2024-10-02T17:26:27Z")
**Posts on this page:** 3
**Page:** 1

<div class="post-metadata">

### Author: ![Dimitrios\_Samaras](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/dimitrios_samaras/32/78102_2.png) [@Dimitrios\_Samaras](https://discourse.slicer.org/u/Dimitrios_Samaras)
#### Post date: [October 2, 2024, 5:26pm UTC](https://discourse.slicer.org/t/problem-with-glcm-features/39499/1 "2024-10-02T17:26:28Z")

</div>

Greetings,  
I am a 1st year PhD student and I’m currently working on a project involving feature extraction from MRI images for prostate cancer research, and I have some questions. I developed a python script for feature extraction but in the case of glcm features I meet some problems. Most of the GLCM features have values of 0 or 1. I attach to you the my results and the corresponding python script. Can you please help me??  
Thank you a lot in advance

This is the script

 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/5/e/5e9f77e73d2313b2ccc78903cf5fc8a037fcb6a0.png)  
 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/5/5/55261babd75fec9894a40449c4ee5d5ff222553e.png)  
 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/e/0/e08b172d52c325abc0d7cedff763c0f72b4fd3fc.png)  
 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/7/2/723882a3a5f0b382c60aafef4059ffb1fce65aa0.png)  
 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/e/7/e78a2b85ffbfc436d06edd331ac6168932805135.png)

and this are the results

 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/4/4/442f6fb538404b07ffd2d7913632309309cf21ca.png)  
 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/c/b/cb19fcf19961fbbab3e5cea804ba7fd90310986d.png)  
 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/e/6/e6a6470c0c62b88d3629f63ebf69b8a274d1f803.png)

---

<div class="post-metadata">

### Author: ![Dimitrios\_Samaras](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/dimitrios_samaras/32/78102_2.png) [@Dimitrios\_Samaras](https://discourse.slicer.org/u/Dimitrios_Samaras)
#### Post date: [October 6, 2024, 9:49pm UTC](https://discourse.slicer.org/t/problem-with-glcm-features/39499/2 "2024-10-06T21:49:40Z")

</div>

Hello again. Can you please help me?

---

<div class="post-metadata">

### Author: ![wafaela](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/wafaela/32/78829_2.png) [@wafaela](https://discourse.slicer.org/u/wafaela)
#### Post date: [December 11, 2024, 3:42pm UTC](https://discourse.slicer.org/t/problem-with-glcm-features/39499/3 "2024-12-11T15:42:37Z")

</div>

Hi, I was dealing with the same problem when I came across your issue. In my case, I managed to obtain more realistic values when using 3D volumes with no prior normalization. Maybe you can check if it is the case for your input volumes ?  
I hope this helps,  
Kindly,  
Wafae