SPM99
|
SPM2(b)
|
I. Convert Data
A. Convert data to axial format, SPM2b compatible ANALYZE.
|
I. Convert Data
A. Convert data to axial format, SPM2b compatible ANALYZE.
|
II. Pre-processing:
A. Start SPM, go to "fMRI" side.
B. Motion Correction:
Realign (~10 minutes to calculate transforms and then reslice images)
Number of subjects: 1
Num sessions for subject 1: 1
[Select files, all 110 of them]
Reslice interpolation method: sync
Adjust sampling errors? no
C. Spatial Normalization:
Normalize (~10 minutes to calculate one transform and reslice all images)
Image to determine parameters from: meanfMRI_vis_stim_0000.img
Images to write normalized: rfMRI...0-109.img (all realigned images)
Template image(s): EPI_template_mommri.img
(This is a template developed at the Keck Lab which uses a number of scans from our scanner)
Interpolation method: Bilinear
D. Smoothing:
Smooth (~8 minutes)
smoothing {FWHM in mm} 8 8 12
|
II. Pre-processing
A. Start SPM, go to “fMRI side”.
B. Motion correction:
Spatial Pre-Processing->Realign
Num subjects: 1
Num sessions, subj 1: 1
Select Images: select all images for your subject
Coregister only
C. Spatial Normalization:
Spatial Pre-processing->Normalize
Determine Parameters and Write Normalized
Select Template Image(s): [Select the T1.mnc]
Source Image, Subj 1: [Select the ANATOMICAL? Or FIRST IMAGE? ]
Determine Parameters and Write Normalized
Select Template Image(s): [Select the T1.mnc? or epi.mnc?]
Source Image, Subj 1: [Select the ANATOMICAL? Or FIRST IMAGE? ]
Image to write, subj 1: [Select all functional images for subj 1, then click “Done”]
Source Image, Subj 2: [click “Done”]
NOTE: Need to standardize the defaults…
D. Smoothing:
Spatial Pre-Processing->Smooth
Smoothing {FWHM in mm}: 8 8 12
[Select scans: Select all the w*.img files that were created by the normalization procedure then click Done.]
|
III. Set up the SPM model:
fMRI models (~2 minutes to set up model, ~4 minutes to run it)
What would you like to do? specify and estimate a model
number of sessions: 1
select scans for session 1: [select all of the 110 smoothed scans]
interscan interval {secs} 3
number of conditions or trials: 1 [N.B.: Baseline doesn't count!]
name for condition 1? visual stim
SOA: fixed
SOA (scans) for visual stim: 20 [10+10]
time to 1st trial (scans): 10
parametric modulation: none
are these trials: epochs
Select type of response: fixed response (boxcar)
convolve with hrf: yes
add temporal derivatives: no
epoch length {scans} for visual stim:10
interactions among trials (Volterra): no
user specified regressors: 0
remove Global effects: none
high-pass filter? specify
session cutoff period (secs): 126
Low-pass filter? hrf
Model intrinsic correlstations? none
Setup trial-specific F-contrasts? yes
estimate? now
|
III. Set up the SPM model
A. Specify model
Model specification…->FMRI->Design
Interscan Interval {secs}: 3
Scans per session e.g. 64 64 64: 110
Specify Design in: scans
Select Basis Set: hrf
Model Interactions (Volterra): No
Number of Conditions/trials: 1 [N.B. Baseline still doesn’t count.]
Name for condition/trial 1: visual stim
Vector of onsets – visual stim: 10 30 50 70 90
Duration[s] (events = 0): 10
Parametric modulation: none
Session 1-> Other regressors: 0
NOTE: At this point you can “Model Estimation…->Review Model->Design->Design Orthogonality” to see the contrasts as before.
B. Specify Data:
Model specification…->FMRI->data
Select SPM.mat: [Select SPM.mat file (that was automatically created in III.A) in your working directory and click Done.]
Select Scans for Session 1: [Filter->“sw*.img”, Click “all” then click “Done”]
Global intensity normalization…->Remove Global Effects: none
Temporal autocorrelation options->High-pass filter: none
Session cutoff period (secs): 126
Correct for serial correlations: none
C. Estimate Model
Model specification…->Estimate
Select SPM.mat: [Select previously selected SPM.mat file]
|
IV. Results
Select the "SPM.mat" file from you analysis directory
F-contrasts: Session 1: vis stim
mask with other contrasts? no
title for comparison: Session1: vis stim
corrected height threshold: yes
corrected p value: 0.05
The glass-brain view appears.
Click on button "volume" to see summary
|
IV. Results
Inference->Results
Select SPM.mat: [Select previously selected SPM.mat file]
Select Contrasts: [Select “Sess(1):visual stim” then click “Done”]
Mask with other contrast(s): no
Title for comparison: “Sess(1):visual stim”
P value adjustment to control: FEW
P value (family-wise error): 0.05
& extent threshold {voxels}: 0
The glass-brain view appears.
Click on button "volume" to see summary
|
