Dear FSL team,
i am working with a set of scans with the aim of identifying changes in resting state network activity. we are interested to discover how activity in these networks changes when subjects are presented with pictures of several types. the pictures where presented in 2 blocks with 4 possible types of pictures in each block (high-calorie food items, low-calorie food items, non-food objects and blurred pictures), the order of the different pictures changed between blocks and was different for different subjects and visits. i am now interested specifically to access the volumes which were recorded following the food-related pictures only, but since these will be mixed with those relating to the other picture types, i want to splice them out of the total scan and then remerge them so i can then perform ICA and dual regression to identify the state of the resting networks.
since we know the onset times for the different picture types for each individual scan, our idea was to first fslsplit and then, having created a list of vol numbers of interest for each subject, write a script for these vols to be sought and selected out and then sequentially added to a growing new file using fslmerge (taking into account the likely delay in the haemodynamic response function, so we would aim to choose vols happening ?6 seconds after picture onset - our TR is 3 seconds).
does this sound like a reasonable approach? is there a faster or better way?
any help would be appreciated.
thanks
Joe Starke
research assistant
robert steiner MRI/C3NL
imperial college
|