Dear SPMers
I would be grateful for some advice. I have two different protocols with slightly different experimental paradigms which I would like to combine into group maps. I think that two potential hurdles will occur during spatial preprocessing and estimating the design matrix.
Both protocols are for simple visual stimulation fMRI experiments with a block OFF-ON design.
The protocols are as follows:
1) Partial brain acquisition: 11 near-axial EPI slices (5 mm slice thickness x 0.5 mm gap) through the visual cortex. TR = 4 secs. 120 scans acquired over 8 minutes. Each epoch is 20 seconds (5 scans).
2) Whole brain acquisition: 27 EPI slices (3.6 mm slice thickness x 0.4 mm gap). TR = 4 secs. 150 scans acquired over 10 minutes. Each epoch is 20 seconds (5 scans).
I have several queries.
For each protocol the TA is different although the TR is the same. Should this make much difference with the processing and if so, is there any way I can cater for this.
I do not think I will get too many problems with normalisation but any advice/previous experience would be much appreciated. I am able to normalise the partial brain dataset fairly well at an individual level but have to disable brain masking.
I am not sure how to formulate the design matrix to take into account the extra 30 scans acquired for the wholebrain protocol. Although the reasons are different, I suspect that the design matrix may be similar to that for studies that combine different modalities/paradigms (eg combining 2 studies with an AB and BC design). Should I consider the 'experiment type' (full/partial) as just another confound, and if so how do I implement this? I would be very grateful for any tips to combat this problem.
Thank you for your help
Best Regards
Ahmed Toosy
Institute of Neurology
Queen Square
London WC1N 3BG
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|