We essentially split the ~197 direction in half, and the two halves can't have the exact same number of directions due to how they are stored on the scanner, so "part 1" is 98 directions and "part 2 is 99. each is then collected both AP and PA. FYI, each scan is actually 92 diffusion volumes plus 6 or 7 non-diffusion "b=0"
This is probably available elsewhere under the CCF documentation, but the DWI scans are adapted from here: https://www.humanconnectome.org/study-hcp-lifespan-pilot/phase1b-pilot-parameters The way the sequence runs on the scanner, we set a single "maximum" b-value (b=3000), by which each of the diffusion vectors in the table is scaled. The entries that norm to 1 are b=3000, and the vectors that norm to 0.707 are b=1500. Note: 0.707 = sqrt(0.5). For whatever reason, this is how the scanner handles vector magnitudes. -Keith On Wed, Sep 27, 2017 at 2:29 PM, Glasser, Matthew <[email protected]> wrote: > In general one wants to get as many gradient directions as possible. > Perhaps Mike knows the answer to your other question. > > Matt. > > On 9/28/17, 2:59 AM, "[email protected] on behalf of > Jeffrey Spielberg" <[email protected] on behalf of > [email protected]> wrote: > > >Hi all, > > > >I¹m interested in setting up a diffusion protocol similar to the CCF > >protocol and have two questions. First, what¹s the difference between > >the dir98 and dir99 acquisitions (beyond having different vectors)? It > >looks like both sample on 2 shells, use the same b-values, and have about > >50:50 vectors on each shell, so I¹m not clear on why both are needed. > > > >Second, I was playing around with the Caruyer q-space sampling tool and > >noticed that the output differs from the vectors provided as part of the > >CCF protocol. Specifically, for dir98, they match for 1 shell, but the > >vector magnitudes in the other shell have been shortened to .7071. My > >guess is that this is necessary to tell the scanner to use the second > >b-value (e.g., 1500). Is that correct? > > > >Best, > >Jeff > > > > > > > >-- > >Jeffrey M. Spielberg, Ph.D. > >Assistant Professor, Clinical Science > >Department of Psychological and Brain Sciences > >University of Delaware > >Newark, DE 19716 > > > >Office: 307 McKinly Laboratory > >Lab: Suite 405 Wolf Hall > >Phone: 302.831.7078 > >Email: [email protected]<mailto:[email protected]> > >Website: http://sites.udel.edu/jmsp/ > > > > > >_______________________________________________ > >HCP-Users mailing list > >[email protected] > >http://lists.humanconnectome.org/mailman/listinfo/hcp-users > > > _______________________________________________ > HCP-Users mailing list > [email protected] > http://lists.humanconnectome.org/mailman/listinfo/hcp-users > _______________________________________________ HCP-Users mailing list [email protected] http://lists.humanconnectome.org/mailman/listinfo/hcp-users
