Resting State BOLD Signal Change Data Analysis. Correlation Analysis

Submitted by crmartir on

Hi dpabi team!

first of all, thaks for developing such a useful tool as dpabi and for maintaining it as an open source software.

I have a question about conducting correlation analysis using Resting State fMRI BOLD effect (signal change). I am trying to correlate rs-fMRI data with behavioral test results. I have preprocessed the data with SPM and calculated rs-fMRI analysis with DPARSF-A via dpabi. Then, my inquiry regards if it is possible to conduct a covariate analysis using behavioral scores as explanatory variables in rs-fMRI analysis wth DPARSF-A, in order to identify brain regions where the BOLD signal change covaries with the bahavioral data.

And if not, what result files among the ones created in rs-fMRI analysis with DPARSF-A (in the results folder) should I use to carry out the correlation analysis in SPSS (for example, the ROISignals file).

Thanks in advance for the help.

Kind regards,

Cristian.

Hi Yan Chao-Gan,

the point is that I want to correlate the signal change, that is, the difference between two groups in seed-based resting-state functional connectivity fMRI, with the results of a behavioral test. And using Correlation Analysis in DPABI Statistical Analysis module, I should enter the FC Maps, and therefore, I can only enter the FC Maps from one group to correlate whole brain wih the behavioral scores.

And I am not sure if DPABI allow us to do this kind of analysis.

Again, thanks in advance for the quik reply.

All the best,

Cristian.

Hi Cristian,
I don't quite understand "the difference between two groups". If you have a group of patients and a group of controls, how could you correlate the "difference" with behavioral traits of only one group?

Best,

Hi Yan Chao-Gan,

I thought it was possible to correlare the signal difference (differences in resting state functional connectivity) from a previous contrast between two groups with the results from a behavioral test, but maybe I am wrong.

All in all, thanks for the advice, I will try Correlation Analysis in dpabi.

Sincerely,

Cristian.