hide
Free keywords:
Independent component analysis; Algorithm; Artefact correction; Adaptive noise cancellation
Abstract:
In this chapter, we focus on the artefacts that arise in the EEG during the fMRI acquisition process. Functional MRI using echo planar imaging (EPI) sequences involves the application of rapidly varying magnetic field gradients for spatial encoding of the MR signal and radiofrequency (RF) pulses for spin excitation (see the chapter “The Basics of Functional Magnetic Resonance Imaging”). Early in the implementation of EEG–fMRI, it was observed that the acquisition of an MR image results in complete obscuration of the physiological EEG (Ives et al. 1993; Allen et al. 2000). Electromagnetic induction in the circuit formed by the electrodes, leads, patient and amplifier exposed to a time-varying magnetic field causes an electromotive force. Artefacts induced in the EEG by the scanning process have a strong deterministic component, due to the preprogrammed nature of the RF and gradient switching sequence, and therefore artefact correction is generally considered a lesser problem than pulse-related artefacts (see the chapter “EEG Quality: Origin and Reduction of the EEG Cardiac-Related Artefact”). According to Faraday’s law of induction, the induced electromotive force is proportional to the time derivative of the magnetic flux (summation of the magnetic field perpendicular to the circuit plane over the area circuit), dΦ/dt, and can therefore reflect changes in the field (gradient switching, RF) or in the circuit geometry or position relative to the field due to body motion (Lemieux et al. 1997). Therefore, the combination of body motion with image acquisition artefacts can lead to random variations that represent a real challenge for artefact correction.