Dear FSL community,
I am using an ROI based method to analyze resting state fMRI data. I have obtained the time series information for my ROIs in the posterior cingulate, white matter, CSF, and whole brain and verified that the ROIs are correctly placed. Most of the resulting design matrices look fine, but some appear to have volumes in which there is a "bump" in the time series, fairly synchronized across all the ROIs. When paging through the actual scan, nothing looks unusual. I have attached a few screen captures so you can see what I mean. This is my first time using an ROI method to analyze resting state fMRI data. Is it typical to see this type of thing in the design matrix? Is this grounds for throwing the subject out? Does anyone know what might cause it? I want all my quality control to occur before actually viewing the final network results so as not to bias my results, so I don't yet know what the DMN looks like for these. Thank you for any thoughts you may have on this.
Best,
Meredith
|