# Bone Segmentation for MRI scans
**URL:** https://discourse.slicer.org/t/bone-segmentation-for-mri-scans/36381
**Category:** Support
**Tags:** segmentation, mri, segmentations, segment-editor, musculoskeletal
**Created:** [May 25, 2024, 12:03am UTC](https://discourse.slicer.org/t/bone-segmentation-for-mri-scans/36381 "2024-05-25T00:03:58Z")
**Posts on this page:** 5
**Page:** 1
### Author:  [@BDA](https://discourse.slicer.org/u/BDA)
#### Post date: [May 25, 2024, 12:03am UTC](https://discourse.slicer.org/t/bone-segmentation-for-mri-scans/36381/1 "2024-05-25T00:03:58Z")
Operating system: windows
Slicer version: 5.6.2
Expected behavior: creating 3D bone models from MR images
Actual behavior:
I want to create 3D bone models from MRI scans…I have MR images of the shoulder and knee region in DICOM format, but I couldnt make any segmentation of the humerus, tibia and femur bones with the “threshold” button as in CT and create an image in a three-dimensional window. How can I segment these bones in MR images? I would appreciate your help.
---
### Author:  [@muratmaga](https://discourse.slicer.org/u/muratmaga)
#### Post date: [May 26, 2024, 5:00am UTC](https://discourse.slicer.org/t/bone-segmentation-for-mri-scans/36381/2 "2024-05-26T05:00:19Z")
MR is not a good modality to visualize bones. They usually appear dark (since they don’t have a lot of water). For good visualization/segmentation of bones, you need CT scans.
---
### Author:  [@BDA](https://discourse.slicer.org/u/BDA)
#### Post date: [May 26, 2024, 2:57pm UTC](https://discourse.slicer.org/t/bone-segmentation-for-mri-scans/36381/3 "2024-05-26T14:57:22Z")
I have seen articles where three-dimensional segmentation of bones was done with 3D Slicer from radial MR images and typical MR images. But I think special shooting is required for this. I wondered if it could be done with normal standard typical MR images in the 3D slicer program.
---
### Author:  [@BDA](https://discourse.slicer.org/u/BDA)
#### Post date: [May 28, 2024, 11:03am UTC](https://discourse.slicer.org/t/bone-segmentation-for-mri-scans/36381/4 "2024-05-28T11:03:54Z")
I have seen some articles suc as “Segmentation of the proximal femur in radial MR scans using a random forest classifier and deformable model registration”, “A 3D active model framework for segmentation of proximal femur in MR images” I wondered if it could be done with standard typical MR images in the 3D slicer program.
---
### Author:  [@ThomasVanParys](https://discourse.slicer.org/u/ThomasVanParys)
#### Post date: [May 29, 2024, 11:56am UTC](https://discourse.slicer.org/t/bone-segmentation-for-mri-scans/36381/5 "2024-05-29T11:56:04Z")
For your research, I would personally recommend obtaining CT scans for bones, which will yield a better threshold. It can be done ([[https://web.stanford.edu/class/ee368/Project\_Autumn\_1516/Reports/Migimatsu.pdf](https://web.stanford.edu/class/ee368/Project_Autumn_1516/Reports/Migimatsu.pdf)]), but MRI is primarily for soft tissue and fluid analysis.