# PET/CT image registartion

**URL:** <https://discourse.slicer.org/t/pet-ct-image-registartion/30283>\
**Category:** Support\
**Tags:** registration\
**Created:** [June 28, 2023, 2:24pm UTC](https://discourse.slicer.org/t/pet-ct-image-registartion/30283 "2023-06-28T14:24:27Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![Sbeta](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/sbeta/32/66238_2.png) [@Sbeta](https://discourse.slicer.org/u/Sbeta)\
**Post date:** [June 28, 2023, 2:24pm UTC](https://discourse.slicer.org/t/pet-ct-image-registartion/30283/1 "2023-06-28T14:24:27Z")

</div>

Hi, I am a learner. I have a 512x512x74 CT image with 0.3848x0.3848x3 mm voxel size and PET image,135x135x74 with 3x3x3mm voxel size. Please advise me on the PET/CT registration process with different dimensions and voxel sizes. I tried Resample volume but gives an error. Thank you.

---

<div class="post-metadata">

**Author:** ![pieper](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/pieper/32/8_2.png) [@pieper](https://discourse.slicer.org/u/pieper)\
**Post date:** [June 28, 2023, 8:07pm UTC](https://discourse.slicer.org/t/pet-ct-image-registartion/30283/2 "2023-06-28T20:07:50Z")

</div>

If these were from a fairly modern PET/CT scanner, and you loaded via dicom and the headers were correct, then the images should have loaded in the correct patient space, no resampling needed.

---

<div class="post-metadata">

**Author:** ![Sbeta](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/sbeta/32/66238_2.png) [@Sbeta](https://discourse.slicer.org/u/Sbeta)\
**Post date:** [June 29, 2023, 12:36am UTC](https://discourse.slicer.org/t/pet-ct-image-registartion/30283/3 "2023-06-29T00:36:34Z")

</div>

Thank you ! The original PET images 256x256x74 with voxels size 1.59095x 1.59095 x74 were resized and convolved using Matlab then converted to nrrd format to add in the 3D slicer.

---

<div class="post-metadata">

**Author:** ![pieper](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/pieper/32/8_2.png) [@pieper](https://discourse.slicer.org/u/pieper)\
**Post date:** [June 29, 2023, 9:32am UTC](https://discourse.slicer.org/t/pet-ct-image-registartion/30283/4 "2023-06-29T09:32:23Z")

</div>

It’s likely that the matlab step lost the geometry information. You’ll need to trace the calculation and reconstruct the correct origin, spacing, and directions to make the volumes line up again.

---

<div class="post-metadata">

**Author:** ![Sbeta](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/sbeta/32/66238_2.png) [@Sbeta](https://discourse.slicer.org/u/Sbeta)\
**Post date:** [June 29, 2023, 12:25pm UTC](https://discourse.slicer.org/t/pet-ct-image-registartion/30283/5 "2023-06-29T12:25:10Z")

</div>

Thank you. Yes, I have considered origin, direction and voxel size.

---

<div class="post-metadata">

**Author:** ![Sbeta](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/sbeta/32/66238_2.png) [@Sbeta](https://discourse.slicer.org/u/Sbeta)\
**Post date:** [July 16, 2023, 1:06am UTC](https://discourse.slicer.org/t/pet-ct-image-registartion/30283/6 "2023-07-16T01:06:33Z")

</div>

![Screenshot 2023-07-15 174450](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/5/a/5a7bd7c1560a6c6c64d18ab40d4cadc0f9a7a669.png)

 ![Screenshot 2023-07-15 174526 (1)](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/3/a/3adb707ec4b2df0d00cd297b4115a1ae662fe168.png)  
I have these two images to register, I am not sure if I need to change the origin before landmark registration. After registration, I am intending to see the dose (PET unit Gy) in the selected volume using segment editor and quantification. I will greatly appreciate your help! Thank you.

---

<div class="post-metadata">

**Author:** ![Sbeta](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/sbeta/32/66238_2.png) [@Sbeta](https://discourse.slicer.org/u/Sbeta)\
**Post date:** [July 16, 2023, 4:43am UTC](https://discourse.slicer.org/t/pet-ct-image-registartion/30283/7 "2023-07-16T04:43:19Z")

</div>

I guess I can use Resample scalar/vector/DWI volume module to change the origin through the Manual output parameter.