# An approach to apply the proposed segmentation region from an ML model to DICOM data 

**URL:** <https://discourse.slicer.org/t/an-approach-to-apply-the-proposed-segmentation-region-from-an-ml-model-to-dicom-data/28372>\
**Category:** Development\
**Created:** [March 14, 2023, 11:40am UTC](https://discourse.slicer.org/t/an-approach-to-apply-the-proposed-segmentation-region-from-an-ml-model-to-dicom-data/28372 "2023-03-14T11:40:50Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![mlh](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/mlh/32/70285_2.png) [@mlh](https://discourse.slicer.org/u/mlh)\
**Post date:** [March 14, 2023, 11:40am UTC](https://discourse.slicer.org/t/an-approach-to-apply-the-proposed-segmentation-region-from-an-ml-model-to-dicom-data/28372/1 "2023-03-14T11:40:51Z")

</div>

Hi all,  
I would like to create a new extension which can do the following task:

- Click a point on a DICOM data on Slicer (e.g. a point in a specific human organ)
- Get that point and image, then pass them to the ML model. The model will output a proposed region for segmentation.
- Pass the model’s output to Slicer and segment the region on Slicer.

I highly appreciate if you guys can recommend an approach to do those task. 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:** [March 14, 2023, 1:17pm UTC](https://discourse.slicer.org/t/an-approach-to-apply-the-proposed-segmentation-region-from-an-ml-model-to-dicom-data/28372/2 "2023-03-14T13:17:59Z")

</div>

Sounds reasonable. You should review MONAI Label and TotalSegmentator to see what’s already possible. Then read the programming tutorials and read through the script repository.