# About displaying MNI atlas on patient head MRI

**URL:** <https://discourse.slicer.org/t/about-displaying-mni-atlas-on-patient-head-mri/23759>\
**Category:** Support\
**Created:** [June 7, 2022, 9:54pm UTC](https://discourse.slicer.org/t/about-displaying-mni-atlas-on-patient-head-mri/23759 "2022-06-07T21:54:53Z")\
**Posts on this page:** 9\
**Page:** 1

<div class="post-metadata">

**Author:** ![Lin](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lin/32/15559_2.png) [@Lin](https://discourse.slicer.org/u/Lin)\
**Post date:** [June 7, 2022, 9:54pm UTC](https://discourse.slicer.org/t/about-displaying-mni-atlas-on-patient-head-mri/23759/1 "2022-06-07T21:54:53Z")

</div>

Hi I’m new to use slicer. Can someone point me a bit on how to display the atlases (eg. MNI 2009 atlas or others used by Lead DBS) on patient space MRI images? I tried by dragging the atlas onto slicer, but the center is not aligned, or its’ not in correct location.

 ![IMG_7832](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/c/d/cd0a8f4e0d7daa00fbd6093d6743b4d4ed504249.jpeg)

I tried to google for methods but did not find clear explanation. If I have run Lead DBS normalization beforehand, is it possible to use the results? Thanks for any suggestions.

---

<div class="post-metadata">

**Author:** ![simonoxen](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/simonoxen/32/66398_2.png) [@simonoxen](https://discourse.slicer.org/u/simonoxen)\
**Post date:** [June 8, 2022, 9:48am UTC](https://discourse.slicer.org/t/about-displaying-mni-atlas-on-patient-head-mri/23759/2 "2022-06-08T09:48:28Z")

</div>

Hi Lin, you can import the Lead-DBS normalization results as transforms in Slicer. You can use the inverse transform (MNI to patient) to transform the atlas models into patient space.

---

<div class="post-metadata">

**Author:** ![simonoxen](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/simonoxen/32/66398_2.png) [@simonoxen](https://discourse.slicer.org/u/simonoxen)\
**Post date:** [June 9, 2022, 7:24am UTC](https://discourse.slicer.org/t/about-displaying-mni-atlas-on-patient-head-mri/23759/4 "2022-06-09T07:24:56Z")

</div>

Hi, the glanatComposite is the transform. There should also be a glanatInverseComposite (weird - perhaps an error from lead).

You can use the glanatComposite, import it as transform in Slicer, and invert it in the transforms module. Then apply it to the models.

---

<div class="post-metadata">

**Author:** ![Lin](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lin/32/15559_2.png) [@Lin](https://discourse.slicer.org/u/Lin)\
**Post date:** [June 9, 2022, 10:14am UTC](https://discourse.slicer.org/t/about-displaying-mni-atlas-on-patient-head-mri/23759/5 "2022-06-09T10:14:21Z")

</div>

Thanks a lot, man. I’ll take a try and get back to you soon.

---

<div class="post-metadata">

**Author:** ![simonoxen](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/simonoxen/32/66398_2.png) [@simonoxen](https://discourse.slicer.org/u/simonoxen)\
**Post date:** [June 9, 2022, 3:31pm UTC](https://discourse.slicer.org/t/about-displaying-mni-atlas-on-patient-head-mri/23759/7 "2022-06-09T15:31:54Z")

</div>

The normalization transforms from lead use the anat\_t\*.nii as reference.

If you center the volumes in slicer, you should take into account this transform.

I’d recommend to keep track of the transforms you apply to the images and you can use the transforms hierarchies in slicer to apply transforms to transforms, and so on.

You could also register the anat\_t\* from lead folder to your centered volume. And then use the resulting transform together with the glanat composite. This is probably easier.

This is a bit complex, and requires some knowledge of how images are represented in space, and how slicer manages this. I’d recommend to look a bit into this.

---

<div class="post-metadata">

**Author:** ![Lin](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lin/32/15559_2.png) [@Lin](https://discourse.slicer.org/u/Lin)\
**Post date:** [June 26, 2022, 2:39pm UTC](https://discourse.slicer.org/t/about-displaying-mni-atlas-on-patient-head-mri/23759/8 "2022-06-26T14:39:49Z")

</div>

Hi Simon, thanks for providing the two methods. I tried both but only did it correctly with the second way.  
I’ll keep on trying.  
Many thanks.

---

<div class="post-metadata">

**Author:** ![Lin](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lin/32/15559_2.png) [@Lin](https://discourse.slicer.org/u/Lin)\
**Post date:** [July 1, 2022, 2:28am UTC](https://discourse.slicer.org/t/about-displaying-mni-atlas-on-patient-head-mri/23759/9 "2022-07-01T02:28:11Z")

</div>

Hi Simon, I’ve aligned the atlas, with your help.  
Now I need to do with the Warp module. Can you please tell me a bit about how to use it?  
See the pic attached, what should I use as displacement field? Do I need to create one? I used Elastix to register anat\_t1 with T1 image in slicer, but didn’t get grid transforms. Can you give me some clue on how to proceed? Thanks a lot.

---

<div class="post-metadata">

**Author:** ![Lin](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lin/32/15559_2.png) [@Lin](https://discourse.slicer.org/u/Lin)\
**Post date:** [July 1, 2022, 2:33am UTC](https://discourse.slicer.org/t/about-displaying-mni-atlas-on-patient-head-mri/23759/10 "2022-07-01T02:33:35Z")

</div>

![SlicerApp-real_uo3VgYiSS8](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/e/3/e328f356a6208a91503787dab011b360fa36c6e2.jpeg)

---

<div class="post-metadata">

**Author:** ![simonoxen](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/simonoxen/32/66398_2.png) [@simonoxen](https://discourse.slicer.org/u/simonoxen)\
**Post date:** [July 1, 2022, 7:57am UTC](https://discourse.slicer.org/t/about-displaying-mni-atlas-on-patient-head-mri/23759/11 "2022-07-01T07:57:15Z")

</div>

> [@Lin](#):
>
> what should I use as displacement field?

You’d generally use the normalization output. Or any non linear transform you want to modify.

See [SlicerNetstim/WarpDrive at master · netstim/SlicerNetstim · GitHub](https://github.com/netstim/SlicerNetstim/tree/master/WarpDrive) and [https://netstim.gitbook.io/leaddbs/appendix/warpdrive](https://netstim.gitbook.io/leaddbs/appendix/warpdrive) for some more info. Still working on warpdrive to make it more user friendly.