CSF Signal as a Complex-Valued RETROICOR Regressor Removes Unwanted Physiological Signal and Increases the Accuracy of Spatial Correlation in Complex-Valued fMRI
Mary Kociuba1 and Daniel Rowe1,2
1Mathematics, Statistics, and Computer Science, Marquette University, Milwaukee, WI, United States, 2Biophysics, Medical College of Wisconsin, Milwaukee, WI, United States
Discarding the phase
component of the time-series removes relevant biological information from a
complex-valued signal. Although, commonly implemented retrospective image
correction techniques fail to account for physiological artifacts in both the
magnitude and phase components of the time-series. Using the CSF signal, observed during the
data acquisition, as a complex-valued regressor increases the statistical power
of fMRI analysis, through reducing unwanted physiological variability in the
complex-valued signal of interest. The improved performance of implementing the
complex-valued image correction methods is demonstrated with a comparison of
magnitude-only and complex-valued spatial correlations.
This abstract and the presentation materials are available to members only;
a login is required.
