 |
 |
Hi again Stefano. Saad has kindly run thalamic segmentation with and without targets as termination masks...  On 12 Mar 2008, at 20:25, Tim Behrens wrote: > Hi - they were not, in this paper. > > In my opinion the size of the cortical masks in this study would > mean the difference would be approximately FA. > > it is worth noting that once a single streamline is counted as > passing through one cortical mask, it is never re-counted to the > same mask, so this is not an issue. > > The only difference made by including the same masks as targets and > terminations is that, if this is done, no streamline can ever be > allocated to more than one mask. If termination masks are not > included then this is of course possible. However, this will happen > to a very small minority of pathways if the target masks are very > big and non-overlapping. > > > This segmentation is replicable across all sorts of different > settings and across different tractography algorithms - I would be > amazed if such a small change made a difference. > > > Cheers > > T > > > > On 12 Mar 2008, at 20:10, Marenco, Stefano (NIH/NIMH) [E] wrote: > >> I have a question for Tim Behrens regarding the 2003 paper on >> thalamic segmentation. >> Were the cortical masks used also as termination masks in that paper? >> Do you think that prescribing the cortical masks as termination >> masks would make much of a difference in terms of thalamic >> segmentation? >> My concern is that, even with loopcheck, if the seed and target >> roi are fairly large, as in this case, a lot of recursive tracts >> could be created, possibly changing the attribution of a thalamic >> voxel (with find the biggest) from one mask to another. Should I >> bother about this or not in your opinion? >> >> Stefano Marenco, MD >> NIMH/GCAP >> Building 10, room 4S235 >> 10 Center Drive >> Bethesda, MD 20892 >> tel 301 435-8964 >> fax 301 480-7795 >> email: [log in to unmask] >> >