[MINC-users] fMRI signal extraction -help request

[Penhune Lab]

Greetings,

I have been having some problems creating percent BOLD signal change plots
for my contrasts and wondered if someone could point me in the right
direction.

I have an experiment with two conditions, L and I, and i am interested in
the learning of L across four fMRI runs.
L is a motor sequence; I is a motor baseline.
GOAL: I intend to use signal extractions to help identify the changes in
activity which occur during learning.

How I understand that signal extractions work:
1) Extract voxel values from _ef file of the L-I contrast
2) extract voxel values from _ef file of the I contrast
3) divide the two and multiply by 100 to give signal change of L relative to
I ( 100*(L-I)/abs(I) )
4) plot across the four runs, see how particular areas change with learning
of L

I did this using the extract.m function (with all of the proper coordinate
transforms) but ran into a bit of strangeness. Before multiplying by 100,
the (L-I)/I values normally vary between .01 and 4 or 5, with occasional
voxels giving me values of over 100 or more (both negative and positive
numbers)! Giving me 'normal' signal change variation of 1-500%, which
doesn't seem too normal... As far as I can tell, this was caused by I
contrast values which were very close to zero. This didn't really seem
reasonable, so I took some advice and tried to detrend my data before
specifying the conditions ( i.e., I ran the raw files through fmrilm with
trends= [3 1 1] and without specifying a design matrix). The intent was to
then run fmrilm with the design matrix and contrast specifications to
produce my contrast files, then run the extractions on them (the theory
being that this would be a better set of files to do the extractions on).
Unfortunately I was unable to test this as apparently there is a limit on
the number of frames which you can process in this way - my experiment
contains 265 frames and the limit appears to be 160 (emma gives an error).

So, I have two major questions. Is my understanding of extractions correct?
How can I extract the values properly, such that I can reach my goal?

Thank you in advance!
Chris

-- 
Laboratory for Motor Learning and Neural Plasticity
Directed by Dr. Virginia Penhune
Department of Psychology, Concordia University
SP-A 244, 7141 Sherbrooke St. W
Montreal, QC  H4B 1R6  Canada
(514)848-2424 ext. 7567
http://psychology.concordia.ca/fac/penhune/winindex.html