[FieldTrip] fieldtrip Digest, Vol 80, Issue 7

Wed Jul 12 09:35:09 CEST 2017

Dear Elena, John and the rest,

Allow me to chime in a bit here to bring across John’s answer even more clearly. Indeed ft_volumesegment applies some postprocessing to the tissue-probability-maps (TPMs) as extracted with SPM. This involves amongst others some smoothing and thresholding, which may indeed result in the output to not fit very snugly with the underlying anatomy. However, the amount of smoothing and threshold is in principle in your hands, so you can adjust the parameters to your need. Just have a look at the documentation of the function for how it can be adjusted.

Note, also, that I have made some recent changes to ft_volumesegment to allow for SPM-options to be configureable, i.e. you can influence the behaviour of the SPM-based step of the procedure. It is hard to give an optimal set of parameters here as well, because these depend on the quality of the anatomical images (and probably on the spm version used). The default values that are used in ft_volumesegment have been taken from the defaults that SPM uses, but there is no strict reason why these are optimal for all MR images.

Also, I’d like to advertise the fact that I have now implemented full support for using SPM12 for the segmentation, where there’s now the option to use 1) a 6-tissue type TPM-template (for this you need to specify cfg.spmversion = ‘spm12’, along with cfg.spmmethod=‘new’), and 2) you can even postprocess the segmentation with the spm add-on ‘mars’. This can be done when specifying cfg.spmmethod=‘mars’, in combination with cfg.spmversion=‘spm12’.

For whatever it’s worth…

Happy computing!

Jan-Mathijs

J.M.Schoffelen, MD PhD
Senior Researcher, VIDI-fellow - PI, language in interaction
Telephone: +31-24-3614793
Physical location: room 00.028
Donders Centre for Cognitive Neuroimaging, Nijmegen, The Netherlands

> On 09 Jul 2017, at 14:45, RICHARDS, JOHN <RICHARDS at mailbox.sc.edu> wrote:
> 
> Elena
> 
> The ft_volumesegment uses the SPM program that does gm/wm segmentation into probability maps.  In the case of using =  {'brain', 'skull', 'scalp'}; as input, the program sums the PVE for GM, WM and "CSF" (other matter) to create the brain (+ fill holes and some other); then has a formula for finding the skull (dilate brain, or use bone class from segmented MRI), and scalp.
> 
> So the "skullstrip" portion consists of the brain PVEs from the SPM segmentation program.  If that program does not work correctly then the overlap of the brain mask will not match the actual brain--could either not cover all the brain or cover more than the brain.
> 
> I have had some issues with the SPM algorithm, and its use in the ft_volumesegment is not very flexible.  I do the segmentation outside of FT using FSL tools (e.g., bet, bet2, betsurf) and then import the segmented head into FT format; FSL also has an algorithm for "skull stripping" found in its VBM script.  This way I have more control over the initial segmented products.  Additionally, neither the SPM nor the FSL routines work with some pediatric populations, especially infants under 2 years of age. In that case I often substitute the initial MRI template (in FT case, as SPM, they use the MNI segmented head) with an infant MRI template and get much better results.  (e.g., neurodevelopmental MRI database).  At times I have used the SPM algorithms for brain segmenting (GM, WM, CSF) and in this case also the PVEs from an age-appropriate brain improve the results.
> 
> John
> 
> 
> ***********************************************
> John E. Richards
> Carolina Distinguished Professor
> Department of Psychology
> University of South Carolina
> Columbia, SC  29208
> Dept Phone: 803 777 2079
> Fax: 803 777 9558
> Email: richards-john at sc.edu
> HTTP: jerlab.psych.sc.edu
> *************************************************
> 
> 
> -----Original Message-----
> From: fieldtrip-bounces at science.ru.nl [mailto:fieldtrip-bounces at science.ru.nl] On Behalf Of fieldtrip-request at science.ru.nl
> Sent: Sunday, July 9, 2017 6:00 AM
> To: fieldtrip at science.ru.nl
> Subject: fieldtrip Digest, Vol 80, Issue 7
> 
> Send fieldtrip mailing list submissions to
> 	fieldtrip at science.ru.nl
> 
> To subscribe or unsubscribe via the World Wide Web, visit
> 	https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
> or, via email, send a message with subject or body 'help' to
> 	fieldtrip-request at science.ru.nl
> 
> You can reach the person managing the list at
> 	fieldtrip-owner at science.ru.nl
> 
> When replying, please edit your Subject line so it is more specific than "Re: Contents of fieldtrip digest..."
> 
> 
> Today's Topics:
> 
>   1. brain segmentation problem (Elena Orekhova)
> 
> 
> ----------------------------------------------------------------------
> 
> Message: 1
> Date: Sat, 8 Jul 2017 19:11:17 +0200
> From: Elena Orekhova <orekhova.elena.v at gmail.com>
> To: FieldTrip discussion list <fieldtrip at science.ru.nl>
> Subject: [FieldTrip] brain segmentation problem
> Message-ID:
> 	<CAGuwe0JDC-b=_hn6_0sox9_hxA-_+edxbx4KXVG5hUQLuTD+Lg at mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
> 
> Dear Fieldtrip experts,
> 
> I noticed that ft_volumesegment sometimes produces inaccurate results, extending the brain mask outside the brain (see the attached figure).  I used
> 
> mri_orig= ft_read_mri ('T1.mgz');
> ...
> cfg           = [];
> cfg.output    =  {'brain', 'skull', 'scalp'};
> mri_segmented = ft_volumesegment(cfg, mri_orig_realigned);
> 
> Have anybody had such a problem. How can it be resolved?
> 
> Elena
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20170708/6bdd59e7/attachment-0001.html>
> -------------- next part --------------
> A non-text attachment was scrubbed...
> Name: fig1.png
> Type: image/png
> Size: 66721 bytes
> Desc: not available
> URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20170708/6bdd59e7/attachment-0001.png>
> 
> ------------------------------
> 
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
> 
> End of fieldtrip Digest, Vol 80, Issue 7
> ****************************************
> 
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip