How does Analyze Skeleton interpret 'surfaces' of junction voxels?

[Chandler, Kathy]

Hi

I'm using Skeletonize/Analyze Skeleton to analyze soil pores. In many of my samples pores have formed around stones and roots creating spaces that form 'planes' rather than relatively circular 'tubes' going through the soil. When I run the Skeletonize plugin it is creating surfaces of junction voxels, but I'm not exactly sure how Analyze Skeleton is interpreting these. Can anyone advise?

I appear to be getting a lot of branches that have a matching start and end point and no Euclidean distance, but a large actual length. Are these the surfaces of junction voxels and, if so, what is the actual length being measured?

I've also got a lot of very short branches and don't know if these are actual short branches or part of these surfaces.

Regards

Kathy

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Ignacio Arganda-Carreras]

Hello Kathy,

Let me try to answer between your lines.

I'm using Skeletonize/Analyze Skeleton to analyze soil pores. In many of my
> samples pores have formed around stones and roots creating spaces that form
> 'planes' rather than relatively circular 'tubes' going through the soil.
> When I run the Skeletonize plugin it is creating surfaces of junction
> voxels, but I'm not exactly sure how Analyze Skeleton is interpreting
> these. Can anyone advise?
>
>
Those planes are usually hard to skeletonize because if you remove any of
their voxels, you would change the topology of volume (and that's not
allowed in the skeletonization algorithm). Therefore, those plane voxels
remain as part of the skeleton and are classified as "junction" voxels
because they have more than 2 neighbors (see the definitions in the site).


> I appear to be getting a lot of branches that have a matching start and
> end point and no Euclidean distance, but a large actual length. Are these
> the surfaces of junction voxels and, if so, what is the actual length being
> measured?
>

Those seem to be circular branches (start  = end). That would explain why
the Euclidean distance is zero but the actual length is large.


> I've also got a lot of very short branches and don't know if these are
> actual short branches or part of these surfaces.
>
>
The branches are defined as skeleton paths between end points and
junctions, junctions and junctions (with slab voxels in the middle) or from
end points to end points. So I don't think those small branches are part of
your planes.

Your problem looks very close to the bone analysis ones. You should try
sending an e-mail to BoneJ list: http://bonej.org/

I hope this helps!

ignacio


> Regards
>
> Kathy
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>



--
Ignacio Arganda-Carreras, Ph.D.
Seung's lab, 46-5065
Department of Brain and Cognitive Sciences
Massachusetts Institute of Technology
43 Vassar St.
Cambridge, MA 02139
USA

Phone: [phone]
Website: http://bioweb.cnb.csic.es/~iarganda/index_EN.html

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Michael Doube-4]

Hi Kathy,

> I've also got a lot of very short branches and don't know if these
> are actual short branches or part of these surfaces.

If your short branch has to connect to something else to count as a
branch, then perhaps the Euler characteristic would help. It's not
foolproof however, because little loops of noisy pixels look like
branches to the algorithm, in which case preprocessing is required.

http://bonej.org/connectivity

Michael



[RVC Logo - link to RVC Website]<http://www.rvc.ac.uk>    [Twitter icon - link to RVC (Official) Twitter] <http://twitter.com/RoyalVetCollege>     [Facebook icon - link to RVC (Official) Facebook] <http://www.facebook.com/theRVC>     [YouTube icon - link to RVC YouTube] <http://www.youtube.com/user/RoyalVetsLondon?feature=mhee>     [Pinterest icon - link to RVC Pinterest] <http://pinterest.com/royalvetcollege/>

This message, together with any attachments, is intended for the stated addressee(s) only and may contain privileged or confidential information. Any views or opinions presented are solely those of the author and do not necessarily represent those of the Royal Veterinary College (RVC). If you are not the intended recipient, please notify the sender and be advised that you have received this message in error and that any use, dissemination, forwarding, printing, or copying is strictly prohibited. Unless stated expressly in this email, this email does not create, form part of, or vary any contractual or unilateral obligation. Email communication cannot be guaranteed to be secure or error free as information could be intercepted, corrupted, amended, lost, destroyed, incomplete or contain viruses. Therefore, we do not accept liability for any such matters or their consequences. Communication with us by email will be taken as acceptance of the risks inherent in doing so.

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Chandler, Kathy]

Hello

Thanks Ignacio and Michael for helping me with this and my other question relating to Analyze Skeleton. In fact I'd like to take the opportunity to thank all in the ImageJ/Fiji/BoneJ community for the time they put into creating, maintaining and supporting this software. Without it I would have been unable to carry out some of the work that I'm currently doing for my PhD.

Ignacio, I checked my images and found that I did indeed have some objects with a single voxel sitting on the boundaries of the image stack, so that resolves the question regarding differences between particle analyser and analyze skeleton. As for the surfaces of junction voxels, if I've understood you correctly about the circular branches, the branch length in these instances actually describes the number of pixels that make up this surface of junction voxels, is that correct?

Many thanks

Kathy

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Ignacio Arganda-Carreras]

Hello again Kathy,

>
> Ignacio, I checked my images and found that I did indeed have some objects with a single voxel sitting on the boundaries of the image stack, so that resolves the question regarding differences between particle analyser and analyze skeleton. As for the surfaces of junction voxels, if I've understood you correctly about the circular branches, the branch length in these instances actually describes the number of pixels that make up this surface of junction voxels, is that correct?
>

Not exactly. The length are calculated based on the calibration of the
images, so the value does not correspond to number of pixels/voxels.
Maybe Mike can tell you more about those plane surfaces since it might
happen more often in bones.

ignacio

--
Ignacio Arganda-Carreras, Ph.D.
Seung's lab, 46-5065
Department of Brain and Cognitive Sciences
Massachusetts Institute of Technology
43 Vassar St.
Cambridge, MA 02139
USA

Phone: [phone]
Website: http://bioweb.cnb.csic.es/~iarganda/index_EN.html

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html