Hi Brian,
There is no golden rule, it depends on which tracts you are after. The higher the angle the more flexible the propagation will be, but also the higher the chance to get false positives.
Cheers
Stam
On 21 Mar 2013, at 17:06, Brian Stirling wrote:
> Dear awesome FSL experts,
>
> I understand the calculations behind the probtrackx curvature threshold value, but I was wondering what your thoughts are on selecting an optimum threshold value for a 2-Fiber Model (HARDI) for both qboot and bedpostx. It's currently defaulted at 0.2, translating to +/- 80 degs. But I'm partial to selecting a threshold that translates to +/- 90 degs. Is this feasible?
>
> Thanks,
>
> Brian
>
|