# Anatomy of Aorta and Pulmonary arteries

**URL:** <https://discourse.slicer.org/t/anatomy-of-aorta-and-pulmonary-arteries/40564>\
**Category:** Support\
**Tags:** segmentation, segmentations, segment-editor\
**Created:** [December 8, 2024, 1:05pm UTC](https://discourse.slicer.org/t/anatomy-of-aorta-and-pulmonary-arteries/40564 "2024-12-08T13:05:47Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![OmerAfsin](https://avatars.discourse-cdn.com/v4/letter/o/eb8c5e/32.png) [@OmerAfsin](https://discourse.slicer.org/u/OmerAfsin)\
**Post date:** [December 8, 2024, 1:05pm UTC](https://discourse.slicer.org/t/anatomy-of-aorta-and-pulmonary-arteries/40564/1 "2024-12-08T13:05:47Z")

</div>

Hello everyone.  
I’m new here and this is my first post in community. If there is something you find wrong in this post, I can remove it if you contact me.

I am working on a project about the automatic segmentation of great arteries from CT images. I study engineering so that my anatomy background is not very good. I started to use a 3D slicer to segment aort and pulmonary arteries, but anatomy and selecting appropriate segmentation tools are hard for me. Can you suggest sources for these issues?

In addition, I want to ask another thing. I exported the segmentation of a dicom slice in NRRD format as a labelmap and read the NRRD file in Python. When I visualized this, I noticed that it was rotated compared to the segmentation I made in 3D slicer. Is that because of the fact that the Dicom and NRRD file keeps the axis records in different orders?

 ![Ekran görüntüsü 2024-12-05 005552](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/2/6/26a06a7533663cbdd26c4c95263139d295d59b77.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:** [December 8, 2024, 1:12pm UTC](https://discourse.slicer.org/t/anatomy-of-aorta-and-pulmonary-arteries/40564/2 "2024-12-08T13:12:00Z")

</div>

You can segment message vessels automatically using MONAIAuto3DSeg, TotalSegmentator, LungCTAnalyzer extension.

If a software displays an image incorrectly then you can report the error to its developers. If they don’t respond then you can try to fix the issue yourself or use a different software.

---

<div class="post-metadata">

**Author:** ![OmerAfsin](https://avatars.discourse-cdn.com/v4/letter/o/eb8c5e/32.png) [@OmerAfsin](https://discourse.slicer.org/u/OmerAfsin)\
**Post date:** [December 8, 2024, 5:25pm UTC](https://discourse.slicer.org/t/anatomy-of-aorta-and-pulmonary-arteries/40564/3 "2024-12-08T17:25:47Z")

</div>

First of all, thank you for answering. I tried the 3 extensions you mentioned.

MonaIauto3dseg and Totalsegmentator returned very good segmentations than I expected.

The aorta has a understandable anatomy, but **Pulmonary Arteries** is a bit difficult. I couldn’t find a simplified source to define its anatomy. If you have a source about this, can you share it with me?

Before I see your answer, I tried to do it myself and I shared the result I made. With this, I plan to train a 3D U-Net model, and I will give the segmentation of 20 such volumes to the model by matching these slices with the corresponding labelmap, respectively. I really appreciate if you comment.

About Nrrd, I just asked because I was wondered, my goal was not to complain.

Thanks for sharing your time.

 ![segmentation1](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/7/0/703ab82599315bc94c5cc1a64b51b1fb34c65a1f.png)