SCI Seg3D Mailing List

Text archives Help

Re: [Seg3D] Brain Routine CT Ventricle Segmentation

Chronological Thread 
  • From: Jess Tate <>
  • To:
  • Cc:
  • Subject: Re: [Seg3D] Brain Routine CT Ventricle Segmentation
  • Date: Thu, 12 Nov 2015 10:05:28 -0700

You can check out the seg3D tutorial ( for some tips on segmentation strategy.  There is a brain example, and though they don’t show the segmentation of the ventricles, you can use the same strategy from the other examples.  In that example Paul’s suggestion of using threshold and connected component may be the fastest.


On Nov 12, 2015, at 8:32 AM, Paul Holcomb <> wrote:


What kind of resolution and contrast do you have in your images? I imagine a simple thresholding and connected component analysis in Seg3D would be able to pull out the ventricles very quickly. How big are your datasets?

Paul Holcomb

On Thu, Nov 12, 2015 at 9:56 AM Elizabeth Jurrus <> wrote:
Hi Kelly,

I am forwarding your question to our mailing list for suggestions from
our user community.

- liz

-------- Forwarded Message --------

I'm interested in doing manual or automated ventricle Segmentation on
routine brain CT scans. I've been using slicer but was wondering if
there was a way to do this in your Seg3D2 program? Thanks.

Archive powered by MHonArc 2.6.18.

Top of page