A Combined Independent Component Analysis ( ICA ) / Empirical Mode Decomposition ( EMD ) method to infer Corticomuscular

coherence has been recently used to investigate the motor system in humans. Typically this is performed by calculating the coherence between a single EEG electrode and a rectified EMG channel. However, there are strong biological reasons to assume that the cortical to muscular communication is many-to-many as opposed to one-to-one. Here we describe the use of ICA to find linear combinations of EEG channels and EMG channels separately. EMD was then used to determine Intrinsic Mode Functions (IMFs) that estimated the envelope of the EMG ICs. We demonstrate that at least 2 EEG ICs correspond with EMG IC IMFs with much greater significance that the pairwise EEG-EMG comparison. Moreover, the proposed method successfully untangled the ~10Hz and ~30Hz effects of the corticomuscular coupling which are thought to underlie different neural processes. We suggest that the ICA/EMD approach is worthy of further exploration. A. INTRODUCTION – EEG/EMG COUPLING EEG/EMG coherence, whereby simultaneously-recorded EMG and EEG signals are compared in the frequency domain, has been used to infer the coupling between the primary motor cortex and the musculature, the final pathway in the human motor system. This method has demonstrated alterations in estimated coupling as a function of motor task, and in diseases of the motor system, such as Parkinson's disease. The standard approach of estimating the coherence between a single EEG channel and a single rectified EMG channel is fundamentally limited by: Assumed temporal stationarity of the EEG and EMG spectra. However, motor movements are naturally dynamic and non-stationary. The common formulation of coherence relies on pairwise comparisons. Nevertheless the biology suggests that the mapping between the brain and musculature is many-to-many, as opposed to one-to-one [1]. Rectification (taking the absolute value) of the EMG to estimate the envelope. While rectification of the EMG when it clearly consists of individual motor units separated in time may result in the frequency spectrum approaching that of the envelope frequency [2], such a situation rarely occurs in practise with surface EMG during anything more than a minimal contraction – the typical scenario. While newer approaches, such as the wavelet coherence [3] [4] may address issue 1), here we examine the latter two issues. Clinically, the pathological situation of cortical myoclonus, which can be considered the equivalent of a motor cortical command delta function, δ(t), results in several muscles being activated as opposed to individual muscles [5]. Georgopoulos has proposed a linear model where cortical …