BOLD correlates of EMG spectral density in cortical myoclonus: description of method and case report.

Journal article

The recording of electrophysiological data during BOLD fMRI is highly challenging but provides the opportunity to develop a more thorough account of brain function than either modality alone. To develop new techniques in this area has often required the study of pathological electrophysiological measures because such measures can be unusually strong (e.g., epileptic EEG spikes) and hence more easily detectable during fMRI than weaker normal phenomena. In this paper, we have studied pathologically strong EMG signals in a patient with cortical myoclonus. First, we studied the patient outside of scanning: The pathological corticospinal drive to muscle was concentrated over 6-30 Hz so that EMG components at higher frequency could be used to control for non-corticospinal contributions to the EMG during scanning. Additionally, there was very strong EMG-EMG and EMG-EEG coherence in this frequency band. Furthermore, analysis of spectral phase showed that this exaggerated coherence was produced by efferent drive from brain to muscle. Hence, the exaggerated peak in distal muscle EMG power spectrum reflected brain activity producing the efferent drive. Subsequently, we modified equipment and data analysis techniques previously developed for simultaneous EEG-fMRI to record polychannel EMG from distal upper limbs during simultaneous BOLD fMRI. After artefact reduction, the EMG recorded during fMRI retained the essential frequency and phase characteristics of EMG recorded outside of scanning. The BOLD signal was significantly correlated with time-varying 6- to 30-Hz power in a frontoparietal network compatible with the neurophysiological characteristics of our patient and compatible with prior hypotheses explaining the origin of cortical myoclonus.

Authors: Richardson, M; Grosse, P; Allen, P; Turner, R; Brown, P

• Publication status: Published
• Journal: NeuroImage
• Volume: 32
• Issue: 2
• Pages: 558-565
• Publication date: 2006-08-01
• DOI: 10.1016/j.neuroimage.2006.04.183
• EISSN: 1095-9572
• ISSN: 1053-8119
• Language: English
• Keywords: Software; Adult; Image Processing, Computer-Assisted; Electroencephalography; Gyrus Cinguli; Fourier Analysis; Signal Processing, Computer-Assisted; Electromyography; Essential Tremor; Myoclonus; Humans; Image Enhancement; Brain Mapping; Nerve Net; Cerebral Cortex; Muscle, Skeletal; Dominance, Cerebral; Mathematical Computing; Magnetic Resonance Imaging; Male; Statistics as Topic; Oxygen; Artifacts