Correlations between cardiac data and fMRI data as a function of spatial frequency and time

Physiological noise from the experimental subject caused by the nearly periodic oscillations of heart beat and respiration is a prominent source of data contamination in functional magnetic resonance imaging (fMRI) studies. Physiological noise can lead to complex distortions in the data, which can in turn affect the quality of the study's results. As this is a serious problem in nearly all fMRI studies, methods for reducing this type of noise have been previously examined by several investigators. Here we focus on the relationships between heart rate and the fMRI data as a function of spatial frequency and time. The time dimension is introduced in order to consider delays that may be present between the heart beat and its maximum effect on the fMRI data. An understanding of these relationships will aid in the development of an accurate and more effective method to remove the heart beat's effects on the fMRI data.