Research Overview - Dual Regression
A common need for analyses such as ICA for resting-state FMRI is to run a group-average ICA, and then, for each subject, estimate a "version" of each of the group-level spatial maps. Currently, the best way to do this is to use dual regression. This:
- Regresses the group-spatial-maps into each subject's 4D dataset to give a set of timecourses (stage 1)
- Regresses those timecourses into the same 4D dataset to get a subject-specific set of spatial maps (stage 2)
It is then common to compare the spatial maps across groups of subjects to look for group differences, ideally using randomise permutation testing.
If you use dual regression in your research, please make sure that you reference at least one of the articles listed below. For your convenience, we provide example text, which you are welcome to use in your methods description.
"The set of spatial maps from the group-average analysis was used to generate subject-specific versions of the spatial maps, and associated timeseries, using dual regression [Beckmann09,Filippini09]. First, for each subject, the group-average set of spatial maps is regressed (as spatial regressors in a multiple regression) into the subject's 4D space-time dataset. This results in a set of subject-specific timeseries, one per group-level spatial map. Next, those timeseries are regressed (as temporal regressors, again in a multiple regression) into the same 4D dataset, resulting in a set of subject-specific spatial maps, one per group-level spatial map. We then tested for [group differences, etc.] using FSL's randomise permutation-testing tool."
[Nickerson 2017] L. Nickerson, S.M. Smith, D. Öngür, C.F. Beckmann. Using Dual Regression to Investigate Network Shape and Amplitude in Functional Connectivity Analyses. Front Neurosci. 2017; 11: 115. doi: 10.3389/fnins.2017.00115
[Beckmann 2009] C.F. Beckmann, C.E. Mackay, N. Filippini, and S.M. Smith. Group comparison of resting-state FMRI data using multi-subject ICA and dual regression. OHBM, 2009.