# MONAI AutoSeg3D for CBCT (CMF)

**URL:** https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175
**Category:** Support
**Tags:** segmentation
**Created:** [July 3, 2024, 12:27pm UTC](https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175 "2024-07-03T12:27:47Z")
**Posts on this page:** 11
**Page:** 1

<div class="post-metadata">

### Author: ![ssingh](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/ssingh/32/77247_2.png) [@ssingh](https://discourse.slicer.org/u/ssingh)
#### Post date: [July 3, 2024, 12:27pm UTC](https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175/1 "2024-07-03T12:27:47Z")

</div>

I tried MONAI AutoSeg3D auto-segmentation on CMF scan (CBCT).  
I see that the segmentation results appear like as if it was done on a downsampled image.

Please see the snapshot here:

 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/c/0/c0ce1dd5a7840bda6c0471e084ba47fe62b7c0c0.jpeg)

I have used Whole Body segmentation TS2. I think TS2 implies Total Segmentator V2. I Observed a similar result when I used the Total Segmentator extension.

I also see the following log in the AutoSeg3d:

```auto

Using orientation_ras

Using crop_foreground

Using resample with resample_resolution [1.5, 1.5, 1.5]

Running Inference ...

```

So it appears it resamples the image to 1.5x1.5x1.5

Is it possible to run the AutoSeg3D to get a better(finer) segmentation mask result? The extension currently does not provide such a control.

Thanks

---

<div class="post-metadata">

### Author: ![Gauthier](https://avatars.discourse-cdn.com/v4/letter/g/919ad9/32.png) [@Gauthier](https://discourse.slicer.org/u/Gauthier)
#### Post date: [July 4, 2024, 8:26am UTC](https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175/2 "2024-07-04T08:26:45Z")

</div>

Hello,

To my knowledge Monain Auto3DSeg has not been trained on those kind of data for now. You could give a try to DentalSegmentator : [https://discourse.slicer.org/t/new-extension-dentalsegmentator-for-automatic-dental-ct-cbct-segmentation](https://discourse.slicer.org/t/new-extension-dentalsegmentator-for-automatic-dental-ct-cbct-segmentation)

Best,  
Gauthier

---

<div class="post-metadata">

### Author: ![ssingh](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/ssingh/32/77247_2.png) [@ssingh](https://discourse.slicer.org/u/ssingh)
#### Post date: [July 4, 2024, 8:35am UTC](https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175/3 "2024-07-04T08:35:18Z")

</div>

@Gauthier thanks for your reply.  
Yes, I have used DentalSegmentator and AMASSS. They perform better. I wanted to see the performance of this tool as well. Basically, then use the best one.

On Dental Segmentator: Only the extension code is open source, however I could not find the implementation on auto segmentation model training. Is that also available, say if someone wants to extend work on Dental Segmentator?

Regards,  
Sukhraj

---

<div class="post-metadata">

### Author: ![Gauthier](https://avatars.discourse-cdn.com/v4/letter/g/919ad9/32.png) [@Gauthier](https://discourse.slicer.org/u/Gauthier)
#### Post date: [July 4, 2024, 8:41am UTC](https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175/4 "2024-07-04T08:41:33Z")

</div>

Hello,

Thanks for your reply.

The pretrained nnU-Net model for DentalSegmentator is shared here : [DentalSegmentator nnU-Net pretrained model for CBCT image segmentation](https://zenodo.org/doi/10.5281/zenodo.10829674)

Best regards,  
Gauthier

---

<div class="post-metadata">

### Author: ![ssingh](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/ssingh/32/77247_2.png) [@ssingh](https://discourse.slicer.org/u/ssingh)
#### Post date: [July 4, 2024, 8:48am UTC](https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175/5 "2024-07-04T08:48:32Z")

</div>

So essentially, learn how to use and train further a pre-trained model in nn-Unet, is all that is needed to extend Dental Segmentator.

---

<div class="post-metadata">

### Author: ![Gauthier](https://avatars.discourse-cdn.com/v4/letter/g/919ad9/32.png) [@Gauthier](https://discourse.slicer.org/u/Gauthier)
#### Post date: [July 4, 2024, 8:57am UTC](https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175/6 "2024-07-04T08:57:14Z")

</div>

Yes that would be the idea! nnU-Net offers a basic implementation of finetuning: [nnUNet/documentation/pretraining\_and\_finetuning.md at master · MIC-DKFZ/nnUNet · GitHub](https://github.com/MIC-DKFZ/nnUNet/blob/master/documentation/pretraining_and_finetuning.md)

We haven’t used it for now, so I could not tell you more about the performance bumps that coud be obtained. It will probably depend a lot on what is your objective!

---

<div class="post-metadata">

### Author: ![philippepellerin](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/philippepellerin/32/16062_2.png) [@philippepellerin](https://discourse.slicer.org/u/philippepellerin)
#### Post date: [July 6, 2024, 3:15pm UTC](https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175/7 "2024-07-06T15:15:53Z")

</div>

Have you tried the whole head model in Monai AutoSeg3D? You will get the skull, the mandible, the rachis, eyeball, brain, and masticatory muscles.

---

<div class="post-metadata">

### Author: ![ssingh](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/ssingh/32/77247_2.png) [@ssingh](https://discourse.slicer.org/u/ssingh)
#### Post date: [July 7, 2024, 4:28pm UTC](https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175/8 "2024-07-07T16:28:32Z")

</div>

@philippepellerin I can’t find the Head Model int the model list, please see the snapshot:  
 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/d/f/df081363df6dce88ecfb41935f4c47b53611e8a4.png)

Am I missing something? I am using Slicer 5.7.x

---

<div class="post-metadata">

### Author: ![ssingh](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/ssingh/32/77247_2.png) [@ssingh](https://discourse.slicer.org/u/ssingh)
#### Post date: [July 7, 2024, 4:38pm UTC](https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175/9 "2024-07-07T16:38:59Z")

</div>

@philippepellerin Thanks for noticing in detail.

I observed a head model “Whole Head Segmentation” is available in 5.6.2 version of slicer.

 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/c/d/cd397909621f449d41e94c912895b8cfa4a15a7a.png)

---

<div class="post-metadata">

### Author: ![Camila\_Da\_Rocha\_Camp](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/camila_da_rocha_camp/32/78987_2.png) [@Camila\_Da\_Rocha\_Camp](https://discourse.slicer.org/u/Camila_Da_Rocha_Camp)
#### Post date: [December 29, 2024, 10:58pm UTC](https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175/10 "2024-12-29T22:58:34Z")

</div>

It is not performing the “whole head segmentation on my software” function. Maybe you know what’s wrong with the software. I’m using 3D slicer version 5.6.2.  
Can you help me?

---

<div class="post-metadata">

### Author: ![ssingh](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/ssingh/32/77247_2.png) [@ssingh](https://discourse.slicer.org/u/ssingh)
#### Post date: [March 31, 2025, 7:34am UTC](https://discourse.slicer.org/t/monai-autoseg3d-for-cbct-cmf/37175/11 "2025-03-31T07:34:04Z")

</div>

Hello,

Sorry for the late reply.

Please download the latest Slicer 5.8.x and install the MONAI Auto3Dseg extension. You should be able to see the model:

 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/a/e/ae7ae925140a91530b67555e086755dd57a75d43.png)
