NONLINEAR FOURIER TIME-SERIES ANALYSIS FOR HUMAN BRAIN MAPPING BY FUNCTIONAL MAGNETIC-RESONANCE-IMAGING

A non-linear parametric model for brain activation detection by functi onal magnetic resonance imaging (FMRI) is proposed. The effects of a d esigned temporal stimulus on the FMRI signal at each brain location in a 36 x 60 spatial grid are estimated from discrete Fourier transforms of the observed time series at each location. The frequency domain re gression model accommodates unobservable and spatially varying haemody namic response functions through their estimated convolutions with the global stimulus. This approach generalizes an existing method for hum an brain mapping. An experiment to detect focal cortical activation du ring primary visual stimulation demonstrates the usefulness of the met hod.