Dear Erik-Jan,
> Thanks for responding!
>
> > Note that non-brain matter can and should be removed with BET prior to
> > registration. Do you do this? If not, then see if this fixes the problem.
>
> No, I did not remove the skull, as BET fails to do this correctly. What
> happened is that a part of the eye-muscles was retained. If I asked bet to
> remove more, gray matter got wiped out.. I thought I could "correct" this by
> registering the 3d to the template with skull, but now I understand that the
> epi-3d registration might fail as result of this.
>
> I tried registering the epi with the 3d after "betting" them both, and that
> works!
Great - this was what I'd hoped!
> So one question remains: is is possible to improve the "bet-results"?
OK, the situation is even better than you might think.
Essentially you can use the bet-results to determine the registration as
that removes a lot of the distracting elements from the structurals.
It might also remove a little too much or too little, but this small amount
is unlikely to adversely affect the *registration*.
Then, once the registration is determined, you do not have to use the bet-results
for anything else. You can simply apply the registrations to your pre-bet
data, giving you the best of both worlds!
Of course you could still apply the registration to the pre-bet data in order to
perform template masking as you described. That should also work properly and
give you better "bet" results.
Hope that's cleared everything up.
All the best,
Mark
|