EMG signal decomposition is the process of resolving a composite EMG signal into its constituent motor unit potential trains (classes) and it can be configured as a classification problem. An EMG signal detected by the tip of an inserted needle electrode is the superposition of the individual electrical contributions of the different motor units that are active, during a muscle contraction, and background interference. <BR>This thesis addresses the process of EMG signal decomposition by developing an interactive classification system, which uses multiple classifier fusion techniques in order to achieve improved classification performance. The developed system combines heterogeneous sets of base classifier ensembles of different kinds and employs either a one level classifier fusion scheme or a hybrid classifier fusion approach. <BR>The hybrid classifier fusion approach is applied as a two-stage combination process that uses a new aggregator module which consists of two combiners: the first at the abstract level of classifier fusion and the other at the measurement level of classifier fusion such that it uses both combiners in a complementary manner. Both combiners may be either data independent or the first combiner data independent and the second data dependent. For the purpose of experimentation, we used as first combiner the majority voting scheme, while we used as the second combiner one of the fixed combination rules behaving as a data independent combiner or the fuzzy integral with the lambda-fuzzy measure as an implicit data dependent combiner. <BR>Once the set of motor unit potential trains are generated by the classifier fusion system, the firing pattern consistency statistics for each train are calculated to detect classification errors in an adaptive fashion. This firing pattern analysis allows the algorithm to modify the threshold of assertion required for assignment of a motor unit potential classification individually for each train based on an expectation of erroneous assignments. <BR>The classifier ensembles consist of a set of different versions of the Certainty classifier, a set of classifiers based on the nearest neighbour decision rule: the fuzzy <em>k</em>-NN and the adaptive fuzzy <em>k</em>-NN classifiers, and a set of classifiers that use a correlation measure as an estimation of the degree of similarity between a pattern and a class template: the matched template filter classifiers and its adaptive counterpart. The base classifiers, besides being of different kinds, utilize different types of features and their performances were investigated using both real and simulated EMG signals of different complexities. The feature sets extracted include time-domain data, first- and second-order discrete derivative data, and wavelet-domain data. <BR>Following the so-called <em>overproduce and choose</em> strategy to classifier ensemble combination, the developed system allows the construction of a large set of candidate base classifiers and then chooses, from the base classifiers pool, subsets of specified number of classifiers to form candidate classifier ensembles. The system then selects the classifier ensemble having the maximum degree of agreement by exploiting a diversity measure for designing classifier teams. The kappa statistic is used as the diversity measure to estimate the level of agreement between the base classifier outputs, i. e. , to measure the degree of decision similarity between the base classifiers. This mechanism of choosing the team's classifiers based on assessing the classifier agreement throughout all the trains and the unassigned category is applied during the one level classifier fusion scheme and the first combiner in the hybrid classifier fusion approach. For the second combiner in the hybrid classifier fusion approach, we choose team classifiers also based on kappa statistics but by assessing the classifiers agreement only across the unassigned category and choose those base classifiers having the minimum agreement. <BR>Performance of the developed classifier fusion system, in both of its variants, i. e. , the one level scheme and the hybrid approach was evaluated using synthetic simulated signals of known properties and real signals and then compared it with the performance of the constituent base classifiers. Across the EMG signal data sets used, the hybrid approach had better average classification performance overall, specially in terms of reducing the number of classification errors.

Muscular fatigue limits athletic performance as well as activities of daily living that require repetitive or sustained contractile activity. The decrease in force output or inability to maintain a given force level during fatigue occurs as the result of neural and muscle physiological factors. In contrast to muscle fatigue, potentiation is an increase in muscle force following voluntary muscle activity. The simultaneously occurring processes of potentiation and fatigue influence force output. The aims of this research were to investigate parameters used to potentiate muscle via electrical stimulation and voluntary contraction, and to better understand how muscle force is sustained, we studied single motor unit firing patterns and force potentiation following muscular endurance training. In study 1, electrical stimulation trains matched for pulse number of various frequencies and of increasing pulse number at a given frequency were administered to determine the effects of these stimulation parameters and of the force-time integral (FTI) produced during the train on potentiation magnitude. No difference in potentiation magnitude was found across trains of matched pulse number for frequencies of 15, 25, 30 and 50 Hz. Potentiation increased as pulse number increased and there was a positive correlation between potentiation and the FTI. In study 2, we measured maximal potentiation following conditioning contractions (CC) of 25%, 50% and 100% maximal voluntary contraction (MVC) and during a 25% MVC fatigue task, pre-post 8 weeks of muscular endurance training. Results showed significant potentiation for all CC intensities. Potentiation increased as CC intensity increased and CC duration required to achieve maximal potentiation decreased as CC intensity increased. Muscular endurance training resulted in increased maximal potentiation, and potentiation was greater during the fatigue task after training. Potentiation was also correlated to endurance time. In study 3, the effects of muscular endurance training on motor unit firing rates were investigated. There was a small increase in mean motor unit firing rates during the course of the fatigue task after training. This research contributes to our understanding of muscular force production and muscular endurance. The findings suggest that motor unit firing frequency and force potentiation may contribute to enhanced muscular endurance. / text

Spinal cord injury is a devastating neurological disorder partially characterized by a loss of motor function below the lesion. The dramatic loss of activity results in muscle atrophy and slow-to-fast transformation of contractile elements, producing smaller, weaker and more fatiguable muscles. Functional electrical stimulation (FES), has been proposed in order to induce muscular activity and reverse these changes. FES has primarily been applied in the periphery, either at the surface or implanted in or around a nerve or muscle. Although this can excite nervous tissue and produce muscular contractions, these systems often produce reversed recruitment of motor units leading to inappropriate force generation and increased fatigue. We applied intraspinal microstimulation (ISMS) through fine microwires implanted into the spinal cord of rats. Electrical stimulation through these microwires caused contractions of the quadriceps muscles in both acute and chronically spinalized animals. We showed that muscle recruitment is significantly more gradual with ISMS in intact rats compared to that produced by a standard nerve cuff. Our results further showed that this was due to preferential activation of fatigue resistant muscle fibers. Given this more orderly recruitment of motor units by ISMS, we tested the muscle phenotypes produced by ISMS or nerve cuffs after chronic stimulation. Surprisingly, over a 30 day stimulation period the quadriceps muscles chronically activated by either daily ISMS or nerve cuff stimulation underwent similar fast-to-slow transformations in fiber type and functional properties. This indicates that the recruitment order of motor units does not play the only role in determining the muscle phenotype. Other factors such as the total daily time of activity may be critically important to the phenotypic outcome of skeletal muscle. Finally, we demonstrated that quadriceps force recruitment by ISMS was unchanged following the 30 day stimulation period. In addition, 30 days of chronic ISMS did not cause observable damage in the spinal cord beyond that incurred by the implantation of sham microwires. These studies advance our understanding of the force recruitment properties, neuromuscular plasticity and damage incurred by ISMS and move us closer to developing a clinically viable ISMS procedure.

No description available.

Objectives: Based on the analysis of electromyographic (EMG) data muscles are often characterized as normal or affected by a neuromuscular disease process. A clinical decision support system (CDSS) for the electrophysiological characterization of muscles by analyzing motor unit potentials (MUPs) was developed to assist physicians and researchers with the diagnosis, treatment & management of neuromuscular disorders and analyzed against criteria for use in a clinical setting. Methods: Quantitative MUP data extracted from various muscles from control subjects and patients from a number of clinics was used to compare the sensitivity, specificity, and accuracy of a number of different clinical decision support methods. The CDSS developed in this work known as AMC-PD has three components: MUP characterization using Pattern Discovery (PD), muscle characterization by taking the average of MUP characterizations and calibrated muscle characterizations. Results: The results demonstrated that AMC-PD achieved higher accuracy than conventional means and outlier analysis. Duration, thickness and number of turns were the most discriminative MUP features for characterizing the muscles studied in this work. Conclusions: AMC-PD achieved higher accuracy than conventional means and outlier analysis. Muscle characterization performed using AMC-PD can facilitate the determination of “possible”, “probable”, or “definite” levels of disease whereas the conventional means and outlier methods can only provide a dichotomous “normal” or “abnormal” decision. Therefore, AMC-PD can be directly used to support clinical decisions related to initial diagnosis as well as treatment and management over time. Decisions are based on facts and not impressions giving electromyography a more reliable role in the diagnosis, management, and treatment of neuromuscular disorders. AMC-PD based calibrated muscle characterization can help make electrophysiological examinations more accurate and objective.

Objectives: Based on the analysis of electromyographic (EMG) data muscles are often characterized as normal or affected by a neuromuscular disease process. A clinical decision support system (CDSS) for the electrophysiological characterization of muscles by analyzing motor unit potentials (MUPs) was developed to assist physicians and researchers with the diagnosis, treatment & management of neuromuscular disorders and analyzed against criteria for use in a clinical setting. Methods: Quantitative MUP data extracted from various muscles from control subjects and patients from a number of clinics was used to compare the sensitivity, specificity, and accuracy of a number of different clinical decision support methods. The CDSS developed in this work known as AMC-PD has three components: MUP characterization using Pattern Discovery (PD), muscle characterization by taking the average of MUP characterizations and calibrated muscle characterizations. Results: The results demonstrated that AMC-PD achieved higher accuracy than conventional means and outlier analysis. Duration, thickness and number of turns were the most discriminative MUP features for characterizing the muscles studied in this work. Conclusions: AMC-PD achieved higher accuracy than conventional means and outlier analysis. Muscle characterization performed using AMC-PD can facilitate the determination of “possible”, “probable”, or “definite” levels of disease whereas the conventional means and outlier methods can only provide a dichotomous “normal” or “abnormal” decision. Therefore, AMC-PD can be directly used to support clinical decisions related to initial diagnosis as well as treatment and management over time. Decisions are based on facts and not impressions giving electromyography a more reliable role in the diagnosis, management, and treatment of neuromuscular disorders. AMC-PD based calibrated muscle characterization can help make electrophysiological examinations more accurate and objective.

Establishing a valid, reliable, and objective method for determining the number of functioning motor units in a muscle is important clinically, as it would provide a quantitative means of documenting changes in neuromuscular health over time. This thesis addressed the reliability of motor unit number estimates (MUNEs) derived using decomposition-based quantitative electromyography (DQEMG) from the extensor digitorum brevis (EDB) and abductor hallucis (AH) muscles. Additionally, the effect of the mean surface motor unit potential (SMUP) parameter averaging method (arithmetic/ensemble), the size-related parameter used to calculate MUNE (amplitude/area), and the type of SMUP marker editing (automatic/manual) was investigated in terms of MUNE values. Two separate analyses on a single data set collected from twenty healthy subjects on two occasions were conducted. MUNEs were calculated by dividing a size-related parameter (amplitude/area) of the compound muscle action potential (CMAP) by the same size-related parameter of a representative mean SMUP. First, paired t-tests investigated differences in MUNEs calculated using arithmetic and ensemble averaged SMUP parameters. Within- and between-day reliability of the two measurements was established using intra-class correlation coefficients (ICCs), coefficients of variation (CV), mean absolute differences (MAD), and Bland Altman limits of agreement (LOA). Second, MUNEs (using both parameters) derived from automated and manually edited SMUPs were compared. The effect of the size-related parameter and editing type was identified using a two-factor, repeated measures analysis of variance. Reliability was determined as described above. Arithmetic averaged SMUP parameters produced smaller MUNEs than those derived from ensemble averaging (p<0.001). SMUP area produced higher MUNEs than SMUP amplitude (p<0.05), except when using arithmetic averaged parameters in AH. Interaction effects between editing type and size parameter were present in both muscles (F>6.68, p<0.001). Between-day MUNEs had lower CVs and MADs, higher ICCs, and narrower LOAs than within-day MUNEs. MUNEs derived from arithmetic averaged SMUP parameters showed the highest reliability (ICCs>0.91). MUNEs calculated from automated SMUP marker placements were highly correlated (r>0.86) and displayed comparable reliabilities to those derived from manual marker placement (ICCs>0.90). To optimize the reproducibility of MUNEs calculated using DQEMG, while minimizing processing time, between-day automated estimates using arithmetic averaged SMUP amplitude is recommended. / Thesis (Master, Rehabilitation Science) -- Queen's University, 2012-08-30 08:32:06.141

No description available.

The series of experiments comprising this thesis investigate how neural inputs arising from higher motor centres (e.g., the motor cortex) and the periphery are translated into a variety of activation patterns of alpha motoneurones during the performance of various muscle contraction types. The thesis consists of six chapters, with the first chapter providing an introduction to the research program, and the final chapter giving a summary of the main research findings. Chapter 2 to 5 each represent stand-alone scientific works. The study presented in Chapter 2 examined whether the soleus (SOL) H-reflex is modulated during shortening contractions in a manner that has been observed for isometric contractions. It was revealed that no significant correlation was found between the SOL H-reflex and increasing plantar flexion torque during shortening contractions (ρ = −0.07, P = 0.15), while a strong positive correlation was observed for the isometric conditions (ρ = 0.99, P < 0.01). Furthermore, no modulation in the H-reflexes via paired stimuli in voluntary shortening contractions suggested that the level of homosynaptic post-activation depression (HPAD) did not change in response to the varying levels of activation in voluntary shortening contractions. Therefore, Ia-excitatory input is likely to be reduced during shortening contractions at increasing intensities, possibly due to a centrally regulated increase in presynaptic inhibition. The study described in Chapter 3 investigated corticospinal-evoked responses in triceps surae muscles during voluntary contractions at varying strengths. Motor-evoked potentials (MEPs) and cervicomedullary motor-evoked potentials (CMEPs) were elicited in the SOL and medial gastrocnemius (MG) muscles using magnetic stimulation over the motor cortex and cervicomedullary junction during voluntary plantar flexions with the torque ranging from 0 to 100% of a maximal voluntary contraction (MVC). In both SOL and MG, MEP and CMEP amplitudes [normalized to maximal M wave (Mmax)] showed an increase, followed by a plateau, over the greater part of the contraction range with responses increasing from 0.2 to 6% of Mmax for SOL and from 0.3 to 10% of Mmax for MG. It was suggested that increases in the evoked responses from the triceps surae muscle over a greater range of contraction strengths than for upper limb muscles, probably stems from differences in the pattern of motor unit recruitment and rate coding for these muscles, and the strength of the corticospinal input. In Chapter 4, in an attempt to investigate how the recruitment and rate coding of motor unit organisation can affect the responsiveness of gross evoked potentials to artificial excitatory stimuli, a computer simulation was performed based upon a physiologically plausible model of the motoneurone. The simulation revealed that the force level where the evoked response commences to decline corresponds approximately to the upper limit of recruitment of motor units. This observation was consistent no matter whether firing rates for low-threshold units exceed those for high-threshold units. Since the simulated results were consistent with previous observations in both individual (single motor unit) and population (motoneurone pool) terms, the proposed model is physiologically plausible and can be useful to predict the evoked EMG response via artificial stimulation protocols, thereby inferring the underlying neural mechanisms occurring at the motoneurone pool during voluntary movements. The study presented in Chapter 5 determined the recruitment range and discharge behaviours in the SOL motor units, and examined the possible influence of persistent inward currents (PICs) on SOL motor unit recruitment and discharge rates. Forty-two clearly identified motor units from five subjects revealed that soleus motor units are recruited progressively from rest to contraction strengths close to 95% of MVC, with low-threshold motor units discharging action potentials slower at their recruitment and with a lower peak rate than later recruited high-threshold units. This observation is in contrast to the ‘onion skin phenomenon’ often reported for the upper limb muscles. Based on positive correlations of the peak discharge rates, initial rates and recruitment order of the units with the magnitude of the onset-offset hysteresis (i.e., a difference in discharge rate between recruitment and de-recruitment) and not PIC contribution, we conclude that discharge behaviours among motor units appear to be related to a variation in an intrinsic property other than PICs.

At the beginning of this work the physiological nature of electromyographic signal is described. Design of miniature, portable EMG amplifier for surface measurement is presented. Creation of program for data visualization is described, followed by description of practical EMG amplifier realization.