Home

ELSstudy

fMRI analysis for the ELS KIDMID task
Joshua Ryu; jhryu25@stanford.edu

Directories (where things are)

All the data and analysis are stored in oak (//oak-smb-iang.stanford.edu/groups).

Base directory for the data: //oak-smb-iang.stanford.edu/groups/users/lrborch/ELSReward/Data

Subject level:

• Preprocessing: base/[T1 or Tk1]/[subjId]/spm
• GLM in normalized space: base/[T1 or Tk1]/[subjId]/glm_normSpace
  • /contrasts/ folder contains the contrast images.
• GLM in subject space: base/[T1 or Tk1]/[subjId]/glm_nsubjSpace
  • /contrasts/ folder contains the contrast images.
  • /rois/ folder contains the beta values for each rois.

Group level:

• base/[GroupAnalyses]/[Date]/glm_normSpace: contains the t-statistics across the subject beta value contrasts.

Each date refers to the data of the analysis and contains different number of subjects, depending on the inclusion/exclusion criteria.

Codes for analysis:

Diagram of pipeline:

Preprocessing

• SPM: prep_coregSPM.sh; preprocessing_spm.m
  
• Freesurfer: prep_segmentation.sh
  
• Convert freesurfer brain mask to nii: convertMgz2Nii.sh
  

Behavioral analysis:

• behaviorAnalysis: behaviorAnalysis.m
  

ROIs:

• Generate a list of ROIS from freesurfer: roiListGen.m
  
• Extract ROI: ROI_analysis.m;
  

GLM:

GLM in normalized space: GLM_normspace

• preprocess_spm.m
• normalize_betas.m
• extract_ROI_betas.m
• secondlevel.m
• collectBeta_xsubjs.m
• corrMap.m

Set up the codes to run your code on different computers:

Go to setbasepath.m and set the basedir to according to the path for the computer you are running the analysis. The path is going to depend on how you mounted the drives on the computer and where your data is. Keep the file structure consistent with the original.