Portal de Periódicos da CAPES

Identification of Oscillations in Muscle Activity From Surface EMG: Reply to Halliday and Farmer

2010; American Physiological Society; Volume: 103; Issue: 6 Linguagem: Inglês

10.1152/jn.00325.2010

1522-1598

Evangelos A. Christou, Osmar Pinto Neto,

EEG and Brain-Computer Interfaces

Letters to the EditorIdentification of Oscillations in Muscle Activity From Surface EMG: Reply to Halliday and FarmerEvangelos Christou, and Osmar NetoEvangelos ChristouNeuromuscular Physiology Laboratory, Department of Health and Kinesiology, Texas A&M University, College Station, Texas, and Osmar NetoNeuromuscular Physiology Laboratory, Department of Health and Kinesiology, Texas A&M University, College Station, TexasPublished Online:01 Jun 2010https://doi.org/10.1152/jn.00325.2010MoreSectionsPDF (88 KB)Download PDF ToolsExport citationAdd to favoritesGet permissionsTrack citations ShareShare onFacebookTwitterLinkedInWeChat reply: The surface EMG signal is typically a recording of many motor unit action potentials from a muscle. Despite the fact that it is a crude representation of the organization of individual motor unit action potentials, we agree with Halliday and Farmer (2010) that it may contain important information related to motor unit timing. The issue is whether one can accurately extract such information from the interference (raw) or the rectified EMG signal to better understand motor control.Motor unit timing, however, is a broad term. Studies from single motor units provide at least the following three distinct timing-related sets of information (frequency bands). 1) The average motor unit discharge rate (6–12 Hz). 2) The modulation of motor unit discharge from 0 to 1 Hz (Christou et al. 2007; DeLuca 1982), which likely reflects synaptic noise (Matthews 1996). 3) The modulation of motor unit discharge from 13 to 30 Hz (Christou et al. 2007; Farmer et al. 1993), which has been associated with motor unit synchrony (Christou et al. 2007; Farmer et al. 1993). Subsequently, the following question arises: Which motor unit timing information are we trying to extract from the surface EMG signal? There is evidence that both the interference and rectified EMG signals can identify the mean discharge rate using simple simulations (Myers et al. 2003; see also Fig. 1). During more realistic simulations, however, the rectified EMG impairs the identification of the mean motor unit discharge rate (see Fig. 4 from Farina et al. 2004). There is no evidence that we can extract the 0–1 Hz from either the interference or rectified EMG signal. Finally, although there is evidence for peaks from 13 to 30 Hz in the rectified and interference EMG signals, the origins of such peaks remain unclear. Nonetheless, there is evidence that such peaks from the rectified and interference EMG signals are modulated differently with voluntary effort (Neto et al. 2010).Fig. 1.Example of a simulated spike train signal and its interference and rectified normalized power spectra. A: the simulated signal comprised 10 different simulated spike trains with discharge rates ranging from 9 to 11 Hz. In addition to the mean discharge rate, each spike train contained a 13-Hz oscillation in the interspike interval. The signal on the top demonstrates the shape of a single spike, which had 6 ms duration (all spikes had similar durations). Noise was added to the spike train signal (S:N = 35). B: interference and rectified normalized power spectra. Spectra were normalized by the highest peak between 4 and 16 Hz. The interference spectrum showed 2 distinct peaks, one at 8–10 Hz and the other at 13 Hz. The rectified spectrum, however, showed a single distinct peak, which occurred at 10 Hz.Download figureDownload PowerPointThe major concern raised by Halliday and Farmer (2010) in their letter was that the interference EMG will be primarily concerned with reconstructing the shape of the motor unit action potential as opposed to motor unit firing times. Although most power in the interference EMG signal occurs at higher frequencies (100–200 Hz), there is a significant component at lower frequencies (5–60 Hz). We believe that this band contains significant and physiologically relevant information for the following reasons. 1) There is evidence that the 5- to 60-Hz oscillations in the interference EMG signal are not associated with the shape of the motor unit action potential (Farina et al. 2004; Myers et al. 2003). 2). The 5- to 60-Hz oscillations contain information regarding the timing of motor unit action potentials (e.g., mean discharge rate; Farina et al. 2004; Myers et al. 2003; see also Fig. 1). 3) Yao et al. (2007) demonstrated that corticomuscular coherence is similar for the interference and rectified EMG, demonstrating that at least a portion of these oscillations in the interference EMG is coherent with cortical oscillations. 4) We have recently demonstrated that 12- to 60-Hz oscillations in the interference EMG increased with voluntary effort, which provides evidence for the physiological significance of this band during voluntary contractions (Neto et al., unpublished data).Although for simple simulations the interference and rectified EMG may similarly identify the mean discharge rate of motor units (e.g., 5–12 Hz; Myers et al. 2003; Fig. 1), for more realistic signals only the interference EMG accurately captures the mean discharge rate (Farina et al. 2004). Furthermore, only the interference EMG appears to identify oscillations in muscle activity in other bands of interest (e.g., 13–30 or 30–50 Hz; Farina et al. 2004; Neto et al., unpublished data; see also Fig. 1). Finally, because the literature suggests that components of the 5- to 60-Hz oscillations in the interference EMG signal are physiologically relevant and that rectification of the EMG alters such frequencies via a nonlinear transformation (Farina et al. 2004; Neto and Christou 2010; Yao et al. 2007), we would like to argue that EMG rectification may not be an appropriate preprocessing step to identify the oscillatory input to the motor neuron pool.GRANTSThis work was supported by National Institute on Aging Grant R01-AG-031769 to E. A. Christou.DISCLOSURESNo conflicts of interest are declared by the authors.ACKNOWLEDGMENTSPresent addresses: E. A. Christou, Neuromuscular Physiology Laboratory, Department of Applied Physiology and Kinesiology, University of Florida, FL 32511; O. P. Neto, Neuromuscular Physiology Laboratory, Department of Applied Physiology and Kinesiology, University of Florida, 32511 and Department of Biomedical Engineering, Universidade Camilo Castelo Branco, Brazil.REFERENCES De Luca CJ , Erim Z. Common drive in motor units of a synergistic muscle pair. J Neurophysiol 87: 2200–2204, 2002.Link | ISI | Google Scholar Farina D , Merletti R , Enoka RM. The extraction of neural strategies from the surface EMG. J Appl Physiol 96: 1486–1495, 2004.Link | ISI | Google Scholar Farmer SF , Bremner FD , Halliday DM , Rosenberg JR , Stephens JA. The frequency content of common synaptic inputs to motoneurones studied during voluntary isometric contraction in man. J Physiol 470: 127–155, 1993.Crossref | PubMed | ISI | Google Scholar Halliday DM , Farmer SF. On the need of rectification for surface EMG (Letter to the Editor). J Neurophysiol 103: 000–000, 2010.Link | ISI | Google Scholar Matthews PB. Relationship of firing intervals of human motor units to the trajectory of post-spike after-hyperpolarization and synaptic noise. J Physiol 492: 597–628, 1996.Crossref | PubMed | ISI | Google Scholar Myers LJ , Lowery M , O'Malley M , Vaughan CL , Heneghan C , St Clair Gibson A , Harley YX , Sreenivasan R. Rectification and non-linear pre-processing of EMG signals for cortico-muscular analysis. J Neurosci Methods 124: 157–165, 2003.Crossref | PubMed | ISI | Google Scholar Neto OP , Baweja HS , Christou EA. Increased voluntary drive is associated with changes in the common oscillations from 13–30 and 30–60 Hz of interference but not rectified EMG. Muscle Nerve In press.Google Scholar Neto OP , Christou EA. Rectification of the EMG signal impairs the identification of oscillatory input to the muscle. J Neurophysiol 103: 1093–1103, 2010.Link | ISI | Google Scholar Yao B , Salenius S , Yue GH , Brown RW , Liu JZ. Effects of surface EMG rectification on power and coherence analyses: an EEG and MEG study. J Neurosci Methods 159: 215–223, 2007.Crossref | PubMed | ISI | Google ScholarAUTHOR NOTESAddress for reprint requests and other correspondence: E. A. Christou, Texas A&M University, Department of Health and Kinesiology, Neuromuscular Physiology Laboratory, College Station, TX 77843-4243 (E-mail: [email protected]tamu.edu). Download PDF Previous Back to Top FiguresReferencesRelatedInformation Related ArticlesOn the Need for Rectification of Surface EMG 01 Jun 2010Journal of NeurophysiologyCited ByMotor planning perturbation: muscle activation and reaction timeStefan Delmas, Agostina Casamento-Moran, Seoung Hoon Park, Basma Yacoubi, and Evangelos A. Christou11 October 2018 | Journal of Neurophysiology, Vol. 120, No. 4Pilates and Proprioceptive Neuromuscular Facilitation Methods Induce Similar Strength Gains but Different Neuromuscular Adaptations in Elderly Women26 September 2017 | Experimental Aging Research, Vol. 43, No. 5Characteristics of EMG frequency bands in temporomandibullar disorders patientsJournal of Electromyography and Kinesiology, Vol. 31Synchronous EMG Activity in the Piper Frequency Band Reveals the Corticospinal Demand of Walking Tasks6 June 2013 | Annals of Biomedical Engineering, Vol. 41, No. 8Computation and study of the low-frequency oscillation of surface electromyogram recorded in biceps during isometric upper limb contractionAnkle variability is amplified in older adults due to lower EMG power from 30–60HzHuman Movement Science, Vol. 31, No. 6Long-term adaptations differ for shortening and lengthening contractions21 February 2012 | European Journal of Applied Physiology, Vol. 112, No. 11Hand dominance during constant force isometric contractions: evidence of different cortical drive commands15 December 2011 | European Journal of Applied Physiology, Vol. 112, No. 8Age-associated impairement in endpoint accuracy of goal-directed contractions performed with two fingers is due to altered activation of the synergistic musclesExperimental Gerontology, Vol. 47, No. 7Rectification of the EMG is an unnecessary and inappropriate step in the calculation of Corticomuscular coherenceJournal of Neuroscience Methods, Vol. 205, No. 1Cortical entrainment of human hypoglossal motor unit activitiesChristopher M. Laine, Laura A. Nickerson, and E. Fiona Bailey1 January 2012 | Journal of Neurophysiology, Vol. 107, No. 1Greater amount of visual information exacerbates force control in older adults during constant isometric contractions29 July 2011 | Experimental Brain Research, Vol. 213, No. 4Reliability of multivariate causality measures for neural dataJournal of Neuroscience Methods, Vol. 198, No. 2Neural control of the lips differs for young and older adults following a perturbation18 September 2010 | Experimental Brain Research, Vol. 206, No. 3Reply to Boonstra: The Nature of Periodic Input to the MuscleEvangelos A. Christou, and Osmar P. Neto1 July 2010 | Journal of Neurophysiology, Vol. 104, No. 1 More from this issue > Volume 103Issue 6June 2010Pages 3548-3549 Copyright & PermissionsCopyright © 2010 the American Physiological Societyhttps://doi.org/10.1152/jn.00325.2010History Published online 1 June 2010 Published in print 1 June 2010 Metrics