Hi
> I am aware of that I can use different P value to
> threshold the my result of TBSS analysis, but what I am
> really asking for is whether I can further threshold the
> results using cluster size.
Once you're looking at significant results after correction for multiple comparisons using TFCE, you can always focus on the bigger clusters, but the size threshold will be somehow arbitrary. You can alternatively use the cluster based correction (-c and/or -C in randomise) to see if this "reduces" the extent of your results.
> And more importantly, since
> there are different clusters in my result (I know the
> cluster command can provde some information about clusters,
> but that is really not what I am asking for), how can I save
> each of these differnt clusters into seperately image files?
With the cluster command.
cluster -i tfce_corrp -t 0.95 -o my_cluster_image
and then use fslmaths to threshold it appropriately to extract the clusters of interest.
Cheers,
Gwenaelle
|