# Nvidia AI Assisted Annotation with CT images

**URL:** <https://discourse.slicer.org/t/nvidia-ai-assisted-annotation-with-ct-images/20556>\
**Category:** Support\
**Tags:** segmentation, dicom\
**Created:** [November 9, 2021, 10:34pm UTC](https://discourse.slicer.org/t/nvidia-ai-assisted-annotation-with-ct-images/20556 "2021-11-09T22:34:32Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![TaraGh](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/taragh/32/13054_2.png) [@TaraGh](https://discourse.slicer.org/u/TaraGh)\
**Post date:** [November 9, 2021, 10:34pm UTC](https://discourse.slicer.org/t/nvidia-ai-assisted-annotation-with-ct-images/20556/1 "2021-11-09T22:34:32Z")

</div>

Hello, I have a problem in using Nvidia AI-Assisted Annotation for CT images of the head and neck. I want to segment organs at risk with this extension but if I run, I get errors like this images:  
(I’ have installed the 4.11 version of 3D slicer and in using this extension I select clara-pt-brain-mri-segmentation-t1c. I can’t find a model devoted to CT images)

![photo_2021-11-10_01-53-28](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/a/2/a2550d6fe295c32f01a62ebe6ba8e95c1427c568.jpeg)  
 ![Annotation 2021-11-10 020335](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/3/6/361f6bc6a2ef566cc331cd66df281702b4ad5c1b.png)

I’m new to 3D slicer software, please guide me. does exist another solution in the AI-assisted method for my application?

---

<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:** [November 9, 2021, 11:31pm UTC](https://discourse.slicer.org/t/nvidia-ai-assisted-annotation-with-ct-images/20556/2 "2021-11-09T23:31:58Z")

</div>

This model is trained on MRI images therefore it is expected to fail or return with error for CT images.

You can train your own models using [MONAILabel extension](https://github.com/Project-MONAI/MONAILabel#monai-label).

What structures would you like to segment?

---

<div class="post-metadata">

**Author:** ![TaraGh](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/taragh/32/13054_2.png) [@TaraGh](https://discourse.slicer.org/u/TaraGh)\
**Post date:** [November 10, 2021, 8:05am UTC](https://discourse.slicer.org/t/nvidia-ai-assisted-annotation-with-ct-images/20556/3 "2021-11-10T08:05:07Z")

</div>

Thank you a lot.  
I want to segment organs at risk from these CT images, such as left/right eye, lens, optic nerve, optic chiasm, pituitary … overall about 17 structures.

---

<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:** [November 10, 2021, 7:03pm UTC](https://discourse.slicer.org/t/nvidia-ai-assisted-annotation-with-ct-images/20556/4 "2021-11-10T19:03:10Z")

</div>

I’m not aware of a publicly available model exist for such a comprehensive segmentation task, but you can ask the [MONAI community](https://monai.io/community.html) if they know about something like this or get advice for how to create it yourself.

---

<div class="post-metadata">

**Author:** ![TaraGh](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/taragh/32/13054_2.png) [@TaraGh](https://discourse.slicer.org/u/TaraGh)\
**Post date:** [November 10, 2021, 8:56pm UTC](https://discourse.slicer.org/t/nvidia-ai-assisted-annotation-with-ct-images/20556/5 "2021-11-10T20:56:32Z")

</div>

fine, I’ll do it.

Thanks for the helpful advice.