# Segmentation from Landmark Registration images not aligned

**URL:** <https://discourse.slicer.org/t/segmentation-from-landmark-registration-images-not-aligned/17404>\
**Category:** Support\
**Tags:** registration\
**Created:** [May 1, 2021, 4:44pm UTC](https://discourse.slicer.org/t/segmentation-from-landmark-registration-images-not-aligned/17404 "2021-05-01T16:44:36Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![arif](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/arif/32/10690_2.png) [@arif](https://discourse.slicer.org/u/arif)\
**Post date:** [May 1, 2021, 4:44pm UTC](https://discourse.slicer.org/t/segmentation-from-landmark-registration-images-not-aligned/17404/1 "2021-05-01T16:44:36Z")

</div>

Hi;  
I am trying to do segmentation on fusion of 2 image sets(PET-CT and mpMRI). PET-CT is full body and MRI is pelvic. With the help of Landmark Registration module I am able to overlap the two images. I prefer mpMRI is fixed and CT is moving. As mp MRI show great anatomical detail, I need it to be the fixed one. But when I try to segment my ROI on the fusion image, the result of the segmentation in 3d is far off from where I expect it to be. Is there any suggestion on how to overlap the segmentations also.

Best Regards

 ![Unts](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/2/6/2600bf85cdc1fb9eb3b6ae2c4e45c99f3bcad8df.png)

 ![registration](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/d/f/dfca57fee6353887c9e5411f71c4b8c926e29704.png)

---

<div class="post-metadata">

**Author:** ![lassoan](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lassoan/32/13_2.png) [@lassoan](https://discourse.slicer.org/u/lassoan)\
**Post date:** [May 1, 2021, 6:40pm UTC](https://discourse.slicer.org/t/segmentation-from-landmark-registration-images-not-aligned/17404/2 "2021-05-01T18:40:00Z")

</div>

You can apply the same transform to the segmentation as you applied to the underlying image. See instructions [here](https://slicer.readthedocs.io/en/latest/user_guide/modules/transforms.html#apply-transform-to-a-node).