 |
 |
Hi Arsalan, It sounds you are only using one of the two sessions, correct? Then yes, design and input data are fine. You should also include the option -demean, to mean center data and design (that here has just one regressor). You can enter all four 4D files (with four -i), then -corrmod to correct across all. Make sure subjects are in the exact same order across all of them. No need for -corrcon here, because there is only 1 contrast at the group level. All four contrasts from the first level here become modalities. All the best, Anderson On Tue, 28 May 2019 at 19:30, Arsalan M <[log in to unmask]> wrote: > When to use/ correct usage of: -ise -ee -corrmod -corrcon > > Experimental design: > > Tasked based fMRI was acquired pre and post treatment for all patients. > Disease severity was also measured pre and post treatment to use as > covariate of interest in analysis. All patients were preprocessed in FSL > and for each contrast of interest the MNI registered COPE images of each > individual were concatenated into a 4D.nii file > > Pre scan vs study outcome/measure: > > ie. > fslmerge -t Cope1_contrast_1_4D.nii > /SUBJECT_1_MR_01/Study.feat/reg_standard/stats/cope1.nii.gz > /SUBJECT_2_MR_01/Study.feat/reg_standard/stats/cope1.nii.gz > /SUBJECT_3_MR_01/Study.feat/reg_standard/stats/cope1.nii.gz > /SUBJECT_4_MR_01/Study.feat/reg_standard/stats/cope1.nii.gz > /SUBJECT_5_MR_01/Study.feat/reg_standard/stats/cope1.nii.gz > . > . > . > > Cope1_contrast_1_4D.nii ---> (Reward Cue+) > Cope2_contrast_2_4D.nii ---> (Reward Cue-) > Cope3_contrast_3_4D.nii ---> (Non Reward Cue+) > Cope4_contrast_4_4D.nii ---> (Non Reward Cue-) > . > . > . > > > palm > -i /Cope1_contrast_1_4D.nii > -o /Contrast_1 /PreScan_vs_Outcome_Cope_1 > -d /Design_1SampleTTest_PreScan_vs_ Outcome_AgeSexSite.mat > -t /Design_1SampleTTest_PreScan_vs_HRSD_Cov_AgeSexSite.con > -m /MNI152_T1_2mm_brain_mask.nii > -n 5000 > -T > -C 3.1 > -save1-p > > 1) Is the above palm script suitable to measure the activation with > respect to each individual contrast and a covariate of interest or do I > need to enter all contrasts into the palm command preceded by (-i) to > correctly measure the statistical significance of each contrast or can each > contrast be calculated separately and be statistically valid? > > 2) When would it be appropriate to use -ise, -ee, -corrmod, -corrcon? > (I believe that I only have one modality which is fMRI data, but multiple > contrasts that i want to compare with my covariates of interest would I > then have to use -corrcon in the palm command?) > > Please see bellow for .mat and .con file structures. > Thank you so much in advance for your help. > Kind regards, > Arsalan > > ---- > > /Design_1SampleTTest_PreScan_vs_HRSD_Cov_AgeSexSite.con > > /ContrastName1 percent_improvement > /NumWaves 1 > /NumContrasts 1 > /PPheights 1.625641e+00 > /RequiredEffect 1.915 > > /Matrix > 1.000000e+00 > > ---- > > Design_1SampleTTest_PreScan_vs_ Outcome_AgeSexSite.mat > > ]/NumWaves 1 > /NumPoints 121 > /PPheights 1.625641e+00 > > /Matrix > 2.798318e-03 > -6.127491e-02 > -6.127491e-02 > -1.781227e-01 > 2.113003e-01 > 3.185878e-01 > -4.562273e-02 > -1.434005e-01 > -6.656227e-01 > 3.218773e-01 > -4.072894e-01 > 1.899328e-01 > -1.989561e-01 > -4.182543e-01 > -8.784496e-02 > 1.899328e-01 > -3.656227e-01 > 2.593773e-01 > -2.174746e-01 > 2.798318e-03 > -1.550964e-01 > . > . > > ######################################################################## > > To unsubscribe from the FSL list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 > ######################################################################## To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1