tag:blogger.com,1999:blog-3788849199522127516.post3094819513993470960..comments2014-08-21T06:05:45.165-07:00Comments on brain + map + statistics: Processing a diffusion phantom with FSLMeena Manihttp://www.blogger.com/profile/06948060616149646934noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-3788849199522127516.post-19488340101011018632013-08-01T15:47:01.725-07:002013-08-01T15:47:01.725-07:00Thijs, thanks. The Tractometer evaluation at SCIL ...Thijs, thanks. The Tractometer evaluation at SCIL is interesting. I have also (somewhat painstakingly) generated some ground truth data from the diffusion phantom but for an entirely different purpose. I plan to make it public at some point. So that's a "third life" if you will.Meenanoreply@blogger.comtag:blogger.com,1999:blog-3788849199522127516.post-84146138984796615312013-08-01T05:31:03.853-07:002013-08-01T05:31:03.853-07:00Hi Meena,
Concerning the changing of sign to corr...Hi Meena,<br /><br />Concerning the changing of sign to correct the gradient directions: based on the 2D view you gave of the uncorrected phantom, you may just as well conclude that y is ok, and x needs to be replaced by -x. In short, this is because the diffusion measurements are symmetrical in the angular domain; i.e. if the data would only be 2D, then the orientations (x,y) = (-x,-y), and thus also (x,-y) = (-x,y).<br />However, don't forget that the fibercup phantom, even though it tries to present a 2D situation, is actually still a 3D object that has been scanned. So, in order to obtain a truely correct gradient table, you should also consider if the z direction is correct in relation to x and y.<br />Finally, you may also be interested to know that the fibercup phantom started a "second life" in context of the tractometer approach (an attempt at validating tractography pipelines in the context of connectomics): http://scil.dinf.usherbrooke.ca/tractometer .<br /><br />Kind regards,<br />ThijsThijshttp://www.blogger.com/profile/07415266277640077562noreply@blogger.com