Dear SPMlist,
I have a block design of functional MRI study, the scan parameters and
the method to get each individual T-maps are as follows,
The PWS study was a block design, TR=2.27583 seconds, the condition for
successive blocks alternated between VNS and rest, starting with VNS,
the whole functional EPI scan comprises of 15 cycles, each cycle
contained 2 epochs, VNS and rest, VNS took 6 scans(time-points) and rest
took 18 scans(time-points). The number of total scans (time-points) was
15*24=360.
After realign, slice timing, spatial normalization and spatial smoothing
the data, the statistical T and Z-maps were generated in the following
ways,
Click “fMRI model”
Pull down menu: Would you like to… “specify a model and estimate the
parameters
Number of session: 1
Number of subject: 1
Input the data to be analyzed: 360 snartpt4307.img
Interscan interval: 2.27583
Scan per session: 360
Conditions or trials: 1
Name of trials: VNS
Stochastic Design: no
SOA: Fixed
SOA for VNS: 24
Time to first trial(scans): 0
Parameter modulation: none
Are these trials : epochs
Select type of response: fix response (Box-Car)
Convolve with HRF: yes
Add temporal derivatives: no
Epoch length (scans) for VNS: 6
User specific regresors: 6, (spm_load the parameters txt. file
generated during “Realign” process).
Remove Global effects: Scale
High pass filter? Specify
Cutoff Period:109( =2*24*2.27583)
Low-pass filter: hrf
Model intrinsic correlations: none
Results assessments
Click"Results" in the SPM MenuWindow
Select the SPM.mat that was produced and press 'Done'
The contrast manager will appear.
select 'define new contrast':
Type 'positive activation' in the name area (at the top)
Select "t-contrast"
Type a “1 0 0 0 0 0 0” in the contrast area to detect positive
activation (in the middle)
Submit
Ok
select 'define new contrast'
Type 'negative activation' in the name area (at the top)
Select "t-contrast"
Type a “-1 0 0 0 0 0 0” in the contrast area to detect negative
activation
(The contrast is depicted above the design matrix)
Press "OK"
Select the newly defined ‘positive activation’ or ‘negative activation’
contrast & press "done"
Mask with other contrasts – No
P value = 0.05
Number of voxel: 0
(short wait - progress displayed in MatLab command window)
and the positive and negative T-maps were written to spm99.
Question1: the High pass filter Cutoff Period calculated by SPM99 is
114? Based upon SPM99 manual, it should be 109(( =2*24*2.27583), I do
not know why the number is different? The individual T-map and group
T-maps look reasonable when 114 is used, can I use it instead of 109, or
I must use 109?
Question 2: I have 9 subjects' EPI scans, I want to get the groupT-map,
I have tried this way,
Click “fMRI model”
Pull down menu: Would you like to… “specify a model and estimate the
parameters
Number of session:9
then selected the 9 subjects' data separately,
.......
user specified regressor: 54 (I want to remove motion related
activation, each subject had 6 regressors from motion detection
parameters .txt files, the number of regressor=6*9=54)
then input 9 .txt files separately.
After getting the group spm.mat file, what are the correct contrasts to
get the positive T-map?
1 0 0 0 0 0 0;
1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ( number of "1" is
9, and number of "0" =54);
1 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 1
0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0
Question3, Are there any other ways to get the group T-maps from each
individual T-map or EPI data?
Question4, SPM99 T-maps give T value and Z-scores, what does it mean?
it's a T-map or Z-map, or both? when I do the render analysis, do I
still need to convert it the Z-maps? if so, how to determine the degree
of freedom for each individual T-map and group T-map?
Any suggestion is greatly appreciated!
Qiwen
|