# Centerline computation problem - straight lines

**URL:** <https://discourse.slicer.org/t/centerline-computation-problem-straight-lines/32397>\
**Category:** VMTK\
**Tags:** vmtk, vmtk-centerline\
**Created:** [October 24, 2023, 2:39pm UTC](https://discourse.slicer.org/t/centerline-computation-problem-straight-lines/32397 "2023-10-24T14:39:42Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![smoudour](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/smoudour/32/18451_2.png) [@smoudour](https://discourse.slicer.org/u/smoudour)\
**Post date:** [October 24, 2023, 2:39pm UTC](https://discourse.slicer.org/t/centerline-computation-problem-straight-lines/32397/1 "2023-10-24T14:39:42Z")

</div>

Hi all.

I have been working with 3D slicer for a few months now. I am trying to isolate coronary arteries from CCTA scans, get the centerlines and quantify some geometric features I need for further analysis.

As far, I have managed to get some adequate segmentations of the coronaries and succesful **network model - curves - properties**. What I cannot seem to get is accurate **centerlines**. Specifically, while I get an (almost) correct centerline model, the centerline curves constructed are not good. I’m attaching an image for reference. Could someone explain why do I get this result? I have messed with the endpoints (both manually placed and auto detected) and I always get these weird straight lines. When computing the centerline curves, I get this warning:

`[VTK] Warning: In vtkDelaunay3D.cxx, line 526`  
`[VTK] vtkDelaunay3D (00000236E6A111E0): 1 degenerate triangles encountered, mesh quality suspect`

Here is an image of the constructed centerline **tree** :

 ![centerline_problem](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/7/c/7c39882295d564516141ac30ce69628ca56c0df2.jpeg)

As a reference, the **centerline model** is pretty accurate:

 ![centerline_model_OK](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/d/c/dc0c5f93ad9b9f81c8810be9803c4e83b3b77a6b.png)

I’m trying to get the centerline tree geometry quantification because I want to compute angles of bifurcations along other features, and after some research I found out it is better suited for this task (_reference post_ : [Separating centerline metrics](https://discourse.slicer.org/t/separating-centerline-metrics/19173) ).

My questions are:

- why do I get these weird straight lines. How can I access this “mesh quality” error? How can I fix it?
- what is the correct way to place the endpoints if there is some preffered manned that is.
- Does it matter that in the same segmentation also lives the Right Coronary Artery?  
 ![total](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/8/d/8d5915f12468a9b2e0b1f4fb49281545e70bd707.png)
