Spectral analysis of the surface electromyogram as a tool for studying rate modulation: a comparison between theory, simulation, and experiment

Eccentric Versus Concentric Resistance Training to Enhance Neuromuscular Activation and Walking Speed Following Stroke

Background. Impaired voluntary neuromuscular activation of agonist muscles is a primary determinant of weakness and motor dysfunction following stroke. Objective. To determine whether eccentric resistance training (ECC) resistance training is superior to concentric resistance training (CON) resistance training to enhance neuromuscular activation, strength, and walking speed after stroke. Methods. A total of 34 adults poststroke participated in a staged intervention comprising (1) either CON-only or ECC-only resistance training of the paretic leg followed by (2) gait training. Changes in voluntary neuromuscular activation and power were assessed for both the trained paretic and untrained nonparetic legs. Self-selected and fast walking speeds were also assessed. Results. In response to resistance training, the ECC group experienced larger improvements in neuromuscular activation of paretic leg muscles, rectus femoris and vastus medialis ( P < .005), and the largest gains in paretic leg power (+74% for ECC contractions, P < .0001). ECC also had greater cross-education of increased power to the untrained nonparetic leg (12%-14%, P = .006). Over the course of gait training, much of the gain in paretic leg activation in the ECC group was lost, such that the net change in agonist activation was comparable between the CON and ECC groups when the full intervention was completed. Nevertheless, improvement in walking speed postintervention was more prevalent in the ECC than CON group. Conclusion. ECC resistance training was more effective for improving bilateral neuromuscular activation, strength, and walking speed following stroke. Future research should assess whether a longer duration ECC training program can provide further benefit.

Relationships between surface EMG variables and motor unit firing rates

Although surface electromyography (sEMG) is a widely used electrophysiological technique, its physiological interpretation remains somewhat controversial. This study examined the relationship between motor unit firing rates (MUFR) and the root mean square (RMS) amplitude and mean power frequency (MPF) of the sEMG signal in the biceps brachii. Eleven subjects performed maximal isometric elbow flexion while indwelling and sEMG recordings were obtained from the biceps. The RMS amplitude and MPF of the surface signal, and the mean MUFR from the indwelling signal, were calculated over 500 ms epochs. Group means showed a strong MUFR-RMS amplitude relationship (r (2) = 0.91), but a weak MUFR-MPF relationship (r (2) = 0.20). Using all trials, the MUFR-RMS amplitude (r (2) = 0.19) and MUFR-MPF (r (2) = 0.0037) relationships were much weaker. Within individual subjects, the MUFR-RMS amplitude (mean r (2) = 0.13 +/- 0.17) and the MUFR-MPF (mean r (2) = 0.040 +/- 0.041) relationships were also weak. These results suggest that MUFR cannot be predicted from the characteristics of the sEMG signal.

Mandibular physiological tremor is reduced by increasing-force ramp contractions and periodontal anaesthesia

We have previously shown that the application of anaesthesia to periodontal mechanoreceptors (PMRs) dramatically reduces the 6-12 Hz physiological tremor (PT) in the human mandible during constant isometric contractions where visual feedback is provided. This current study shows that during a ramp contraction where force is slowly increased, the amplitude of mandibular PT is almost five times smaller on average than when the same force ramp is performed in reverse, i.e. force is slowly decreased. This smaller tremor is associated with a higher mean firing rate of motor units (MUs) as measured by the sub-30 Hz peak in the multi-unit power spectrum. The decrease in the amplitude of PT following PMR anaesthetisation is associated in some instances with a similar increase in the overall firing rate; however this change does not match the diminution of tremor. The authors postulate that the decrease in mandibular PT during increasing force ramps may be due to a change in the mean firing rate of the MUs. The change in tremor seen during PMR anaesthetisation may in part be due to a similar mechanism; however other factors must also contribute to this.

Periodontal anaesthesia reduces common 8 Hz input to masseters during isometric biting

During isometric contractions of the jaw muscles, oscillations in the rectified masseteric EMG record that are coherent with the mandibular force output are evident at ~8 Hz. We have investigated the load dependence of these oscillations under both force and EMG feedback conditions and the extent to which these oscillations are coupled bilaterally in the jaw muscles. We further investigated the extent to which afferent information arising from the periodontium during biting influenced the extent of ~8 Hz EMG tremor and the bilateral coupling between masseters at this frequency. Using coherence analysis we have shown that a significant load-independent coupling of EMG between the closing muscles of the jaw occurs at ~8 Hz as a result of common ~8 Hz input to the masseters. This common input is significantly reduced when afferent information from the periodontium is blocked. These results suggest that afferent information arising from the periodontium enhances the expression of peripheral tremulous activity, which may be important for optimising the response of the jaw to changes in forces occurring between the teeth.

Physiology and interpretation of the electromyogram

The purpose of this review is to consider some issues in the interpretation of the electromyogram (EMG) and to discuss current areas of controversy regarding use of the EMG. We consider the underlying physiology and origin of the EMG signal and offer an abbreviated discussion of measurement issues and selected factors that affect the characteristics of the EMG signal. We discuss many of the problems affecting interpretation, including normalization, crosstalk, and issues specific to contraction. In the final section, we consider topics of current interest in electromyography, such as muscle fatigue, task specificity, multichannel representations, and muscle fiber conduction velocity. We present, in addition, alternative analysis techniques. This review should interest researchers and clinicians who seek to obtain the valuable information inherent in the EMG while respecting the potential sources of variance and misinterpretation.

Rate modulation of jaw-elevator motor units as revealed from the low-frequency power spectrum of the surface electromyogram in myogenous CMD patients

The firing pattern of the motor units (MUs) in jaw-elevator muscles was studied within a wide range of isometric contraction levels by means of changes in the frequency and broadness of the primary peak in the low-frequency (5-40 Hz) power spectral density function of the surface EMG. EMG was recorded from both masseter and anterior temporal muscles in 11 myogenous CMD patients as well as in 11 gender- and age-matched controls who clenched in intercuspal occlusion under the control of visual feedback at various levels (0.5-67% MVC for the various muscles studied). The EMG was digitized for 12 periods of 1.6 s per condition; the power spectrum was averaged and smoothed for the various clenching levels. Linear regression analysis showed that the positive slope in the peak frequency (PF)/% MVC relationship, a measure of rate modulation of the MUs, did not differ significantly between patients and controls. At a low clenching level, PF was smaller (p < 0.01) for the anterior temporal muscles of the patients, suggesting lower firing rates for a wide range of clenching levels of the patients because of a similar rate modulation for patients and controls. Furthermore, the variance in the slope values was larger (p < 0.05) for the masseteric muscles of the patients, which may be explained by more heterogeneity of the masseteric rate modulation in the patient group. The broadness of the primary peak was smallest at a low clenching level (p < 0.001) for the anterior temporal muscles of the patients, suggesting a more uniform firing rate or more synchronization between MUs.

Myo-electric fatigue manifestations revisited: power spectrum, conduction velocity, and amplitude of human elbow flexor muscles during isolated and repetitive endurance contractions at 30% maximal voluntary contraction

A brief survey of the literature on manifestations of myo-electric fatigue has disclosed a surprisingly sharp conflict between early studies, focusing on neuromotor regulatory mechanisms, and more recent studies which stress the determinant influence of local metabolism and skewed homeostasis. Favoured explanations concerning changes in the electromyographic (EMG) spectrum were synchronization/grouping of motor unit (MU) firing and conduction velocity (CV) decreases of the action potential propagation. The notion of mutual exclusivity interwoven with these theories prompted us to reinvestigate the EMG of moderate level, static endurance contraction. Ten men in their twenties performed isometric elbow flexion (elbow angle 135 degrees) at 30% maximal voluntary contraction (MVC), and the surface EMG of the brachioradialis (BR) and biceps brachii (BB) muscles was recorded. Initially the CV--determined by cross-correlation--was 4.3 m.s-1 (BR) and 4.6 m.s-1 (BB). At exhaustion the CV of the BR muscle had declined by 33%, roughly twice the decrease of the BB CV. Substantially larger relative median frequency (fm) reductions of 50% (BR) and 43% (BB) were found. Simultaneously, the root-mean-square amplitudes grew by 150% (BR) and 120% (BB). All changes during contraction reached the same level of significance (P < 0.001, both muscles). From the largely uniform relative increases in fm and CV during the last 4 min of a 5-min recovery period, variations in CV were suggested to produce equivalent shifts in fm. The gradually increasing discrepancies between relative decreases in fm and CV during contraction presumably reflected centrally mediated regulation of MU firing patterns (notably synchronization). After the 5-min recovery another 11 endurance contractions at 30% MVC were executed, separated by 5-min intervals. The series of contractions reduced the endurance time to one-third of the 153 s initially sustained, while the terminal CV recordings increased by 1.0 (BR) and 0.6 (BB) m.s-1, and the terminal fm increased by 24 (BR) and 14 (BB) Hz. The relative CV decreased in direct proportion of the endurance time and the fm decreases varied with the CV; the findings did not support a causal link between CV decrease (signifying impaired fibre excitability) and the force failure of exhaustion.

Investigation on parametric analysis of dynamic EMG signals by a muscle-structured simulation model

In the analysis of electromyographic (EMG) signals during dynamic movement, we have proposed an estimation algorithm for the time-varying parameters of an autoregressive model. The parameters correspond to less biased time-varying reflection coefficients. We determined the less biased estimation using a locally quasi-stationary model and named these parameters "k parameters." We estimated k parameters up to the fifth order for the surface EMG signals of a masseter muscle during rapid open-close movement of the lower jaw, a ballistic contraction, and fatigue. According to the results, the time courses of the k parameters displayed remarkable properties. In order to study the behavior of k parameters physiologically, we produced a muscle-structured simulation model based on anatomical and physiological data. The simulation results suggested that the behavior of the third parameter is related to the number of active motor units (MU's) at the shallow layer of a muscle. The detailed recruitment mechanism in terms of the MU's types has not yet been solved. Although further study is required, the parametric analysis using k parameters offers a new perspective for evaluation of muscle dynamics during several movements.

Computer analysis of ENG spectral features from patients with congenital nystagmus

Power spectral estimation from electromystagmographic recording of eye position is proposed as a simple digital processing method to quantify both the amplitude and the frequency features of eye motion in patients affected by the oculomotor disorder of congenital nystagmus (CN). Different basic wave shapes clinically identified and studied in the literature are shown to have slightly different power spectra, which can be used to characterize the CN disorder from nystagmus waveforms. This treatment, statistical in nature, does not depend strongly on the detailed structure of each recorded wave shape, thus emphasizing that accurate descriptive analysis of all patient's waveforms characteristics adds little to the comprehension of CN. We show that the power spectral estimation also represents a useful tool in modelling both the CN defect and the non-defective oculomotor system, and in assessing the effect of the surgical treatment of CN through the differences in the power spectra before and after surgery.

Comments on standardization of reflex measurements in human masseter muscle, including silent periods

The following comments are given on the guidelines suggested by Türker (1988) for studies on oral reflexes using the surface EMG of the masseter muscles in man, including the silent period. (i) Attention should not be confined to electrical stimulation since mechanoreceptors in the periodontium, especially in the periodontal ligament, can only be activated by mechanical stimulation of a tooth. Furthermore, different modes of stimulation and stimulation sites do not yield equivalent reflex complexes. Weak and transient mechanical stimulation of a tooth while clenching at a low level of 5 or 10% MVC is a selective model (89%) to study the influence of periodontal receptors. (ii) Statistical criteria should be applied on rectified and averaged EMG records to prevent any subjective bias in the measurements of reflex variables. Whether a period of increased EMG activity is due to clustering of action potentials of motor units firing of which was delayed by a preceding inhibition, or to a real excitatory influence, can be assessed by applying statistical criteria on averaged and subsequently rectified records. (iii) The clenching level should be less than 25% of MVC to avoid muscle fatigue, and to elicit pronounced reflexes with a weak and therefore more selective stimulus. (iv) The stimulus intensity should be much less than six times the threshold if selectivity for mechanoreceptors is desired. Furthermore, periods of increased activity in surface EMG resulting from a weak stimulus are probably due to real excitatory influences and not to clustering of motor unit action potentials after an inhibition. (v) Jaw separation and the use of a force transducer are not always advantageous. If mechanical stimulation is applied to a tooth, clenching in full habitual occlusion causes a complete suppression of the influence of muscle spindles. A shift in activity between the various elevator muscles, and hence a change in activity of the muscle studied, may occur during an experimental session if biting force is used as a feedback. Such a shift is not relevant if feedback is carried out on rectified and low-pass filtered (1.6 Hz) EMG of the muscle studied.

The effects of motor unit synchronization on the power spectrum of the electromyogram

A realistic model for two synchronized motor unit action potential trains (MUAPT) is presented in which the variability of the time difference between corresponding action potentials (hereafter denoted by delay) is taken into account. Specifically, this delay is modeled as a continuous random variable that may assume both positive and negative values. Expressions are derived for the auto- and cross-power spectra of two such trains using their relations with the auto- and cross-correlation functions, respectively, with which they form Fourier transform pairs. The results show that the auto- and the cross-power spectra of two such synchronized MUAPTs differ from the auto- and the cross-spectra of two independent MUAPTs. The contribution of the statistics of the interpulse intervals to one of the auto-power spectra is smaller and the cross-power spectra no longer reduce to a Dirac sigma-function at the origin but are now determined by the other auto-power spectrum and by the Fourier transform of the density function associated with the time difference between corresponding action potentials. As a consequence of this change in the cross-power spectra synchronization leads to an absolute increase of power at low frequencies and to a relative decrease of power at high frequencies. The results are then generalized to electromyograms (EMG) composed of more than just two MUAPTs and illustrated with simulated power spectra with which the theory shows excellent agreement.