I'd be curious to see a screenshot of your result.
Peace, Matt. _____ From: [email protected] [mailto:[email protected]] On Behalf Of Colin Reveley Sent: Wednesday, February 29, 2012 8:48 PM To: [email protected] Subject: Re: [caret-users] segmenting with diffusion maps Matt - another way to do this, that may even be better, is to use as a base a map of how spherical tensors (or the spheroids they define) are. since the GM is more isotropic than WM, all of the GM comes out as highly spherical. Some bits of WM do too. But one may touch that up with maps of how prolate and oblate the tensors are, since they tend to be one or the other in WM, but neither in GM. together, the three measure segment GM and WM very well. FSL outputs some of this, but not all. I am using the westin WS, WL, WP maps outputted from TORTOISE tensor fitting. I don't know what other programs might make those maps. something else must if tortoise is not an easy option. it looks pretty good. they are also great maps for studying the FSL dyads and why they might have the patterns they do. that's how I stumbled on this in fact. hope to be helpful... On 20 February 2012 18:00, <[email protected]> wrote: Send caret-users mailing list submissions to [email protected] To subscribe or unsubscribe via the World Wide Web, visit http://brainvis.wustl.edu/mailman/listinfo/caret-users or, via email, send a message with subject or body 'help' to [email protected] You can reach the person managing the list at [email protected] When replying, please edit your Subject line so it is more specific than "Re: Contents of caret-users digest..." Today's Topics: 1. Re: segmenting with diffusion maps (Matt Glasser) ---------- Forwarded message ---------- From: Matt Glasser <[email protected]> To: "'Caret, SureFit, and SuMS software users'" <[email protected]> Cc: Date: Mon, 20 Feb 2012 09:15:03 -0600 Subject: Re: [caret-users] segmenting with diffusion maps I use the sum of f (i.e. sum of mean_f#samples) from bedpostx. It will still require a fair amount of manual intervention. Peace, Matt. _____ From: [email protected] [mailto:[email protected]] On Behalf Of Colin Reveley Sent: Sunday, February 19, 2012 12:09 PM To: [email protected] Subject: [caret-users] segmenting with diffusion maps Hi - I have a high quality diffusion data set. I do, actually, also have a structural image but it was aquired a few days later. Although it is the same sample, it seems that it cannot be brougtht into register properly using only linear methods (probably because the ex-vivo sample was pushed around a bit by gravity and fluid pressures; there is a bit of an issue with the containers used). I have access to non-linear methods, but they are intensity based and it's very hard indeed to to bring the structural image into register with an image with a different intensity profile while maintaining the intenisties. I'm continuing to try. But - does any one have any advice concerning segmenting one of the very many tensor maps or other DW volumes I have? It seems to me that the "AM" map, which is an output of tortoise that I think is the mean of b0 vols in principle could work with the intensities reversed, although I've not have much luck. I attach an image. One thing is the bias towards thew caudal end. If I fixed that, does anyone think that this or any other DWI acquisition output could be plausible segmented in CARET? IT doesn't matter how much manual labour I'd have to do. It matters only to get an accurate surface in regsiter with the diffusion data. many, many thanks I guess one thing is that if I could get this image to a stage where it might be viable segmented, then it might also be a viable target for intensity based non-linear registration of a t1-like volume (an MTR image in fact, which segment generally well ) Colin _______________________________________________ caret-users mailing list [email protected] http://brainvis.wustl.edu/mailman/listinfo/caret-users
_______________________________________________ caret-users mailing list [email protected] http://brainvis.wustl.edu/mailman/listinfo/caret-users
