Hi,
It is easiest to run BBR through either FEAT or via the epi_reg script.
It is possible to run directly with FLIRT, although we haven't documented all the advanced options yet.
However, if you do want to try running with FLIRT, you do not need to specify the wm coords or the wm normals as these are internally calculated from the segmentation. The options for specifying coords and normals directly is an alternative way of working, which does not require a segmented image. This latter version is much harder to get right (due to coordinate conventions) so I recommend sticking with the segmented image method, and it is also much easier to use either FEAT or epi_reg. Note that FEAT simply calls epi_reg.
All the best,
Mark
On 25 Sep 2012, at 16:07, Klaus Sembritzki <[log in to unmask]> wrote:
> Hi,
>
> You could let FEAT do the registration and have a look the logs.
>
> Cheers,
> Klaus
>
> On Tue, Sep 25, 2012 at 11:35 AM, Alejandro Vicente Grabovetsky
> <[log in to unmask]> wrote:
>> Dear FSL,
>>
>> I'm having trouble figuring out how to use the BBR method since the
>> documentation seems to give conflicting and somewhat confusing info.
>> I got as far as understanding that I need to use the flirt function with bbr
>> cost functions, but one of the inputs for the BBR to work is the white
>> matter segmentation (easy to obtain from FAST/FIRST or segment 8, etc.) and
>> the white matter boundary coordinates and normals (I don't know where to get
>> these). What function can I get these last 2 inputs from?
>>
>> I'd appreciate any help with this,
>> Best!
>> Sasha
>>
>>
>> --
>> Vicente Grabovetsky, Alejandro (Sasha)
>> Postdoctoral researcher
>> Donders Centre for Cognitive Neuroimaging
>> http://www.doellerlab.com
>>
>>
>>
>
|