Combined interactions of respiratory and cardiac signals measured by high-temporal resolution fMRI

Introduction In fMRI studies, BOLD signal changes are often characterized by a low signal-to-noise ratio (SNR). Physiological artifacts such as cardiac and respiratory noise constitute an important confounding factor that may prevent the detection of significant activations. Moreover, the relatively low temporal resolution of standard fMRI acquisitions cannot adequately sample the artifactual signals, which then become aliased in lower frequency bands. Improvements in SNR have been noted after the removal of low-order aliased harmonics of the cardiac and respiratory frequencies, but a proper characterization and removal of the physiological artifacts requires a higher temporal sampling rate to reduce aliasing effects. The current study investigates the effect of physiological noise in MR-encephalography (MREG), an under-sampled parallel fMRI acquisition method with extremely high temporal resolution. This allows the examination of the effect of high-order harmonics of the cardiac and respiratory signals on the fMRI data, as well as interactions between the two types of physiological artifacts.