Thank you for your assistance in the past with some of my questions regarding mrtrix.
I have run this fibre density and cross section pipeline (Fibre density and cross-section - Multi-tissue CSD — MRtrix 3.0 documentation) for a group of patients (N=29) up to Step #23 where one implements the fixelcfestats command for each metric.
- I would like to obtain the FD, FC, and FDC metrics for each participant just as they are (absent an EVs in the design_matrix) so that I can export them and run stastics in another statistics software (R or SPSS). More specifically, it is my understanding that other processing streams (i.e., fMRIPrep) provide a stats folder where one can export txt or .stats files containing the data for volumes for each region and then conduct analyses in a stats software, which would also enable me to use the FD, FC, and FDC metrics as dependent variables. In reviewing other forum topics, this January 2021 post ‘Understanding GLM global intercept’ by Ana Han noted that “when all EVs are zero, the global intercept is the value of FD / FDC / FC” (which makes sense to me). I then thought this would mean I could upload my files.txt document and
design_matrix.txt [just a single column of 29 1s]
contrast_matrix.txt [with just one zero]
and the outputs would provide me with the FD/FC/FDC values I am looking for (as they would be those values when all EVs are 0).
- However, I have doubts about my inference in this regard, b/c regardless of where I run my stats not all of my EVs would be 0 anyway. Then I thought it might be better to have
design_matrix.txt [column of 1s, with a second column indicating mean centered age in years]
contrast_matrix.txt [with now 2 zeroes]
and the outputs would provide me with the FD/FC/FDC values I am looking for.
**But I am still plagued with doubt (despite looking at GLM youtube videos and other GLM references) and would appreciate feedback/suggestions!
- For kicks to test my theory, I created the design_matrix.txt and contrast_matrix.txt files and ran the fixelcfestats commmands for each and did receive.mif files (such as abseffect.mif and beta0.mif among others) output to the respective directories. Since they are .mif files, however, now I cannot figure out what to do with them to export the outputs to run stats outside of mrtrix. I went through the library of commands on the website and could not find one that would let me convert these to .txt or .tsv, etc. mrconvert did not work either. I also tried opening them with mrview but they just show an empty graphics window.
**Any thoughts on exporting or converting these Fd/FC/FDC metrics for use outside of the pipeline?
I appreciate your time, consideration, and assistance. Happy to provide additional information as needed. Have a wonderful day!