Design and Development of EMG Conditioning System and Hand Gesture Recognition Based on Principal Component Analysis Feature Reduction Technique

This chapter discusses design and development of a surface Electromyogram (EMG) signal detection and conditioning system along with the issues of gratuitous spurious signals such as power line interference, artifacts, etc., which make signals plausible. In order to construe the recognition of hand gestures from EMG signals, Time Domain (TD) and well as Autoregressive (AR) coefficients features are extracted. The extracted features are diminished using the Principal Component Analysis (PCA) to alleviate the burden of the classifier. A four-channel continuous EMG signal conditioning system is developed and EMG signals are acquired from 10 able-bodied subjects to classify the 6 unique movements of hand and wrist. The reduced statistical TD and AR features are used to classify the signal patterns through k Nearest Neighbour (kNN) as well as Neural Network (NN) classifier. Further, EMG signals acquired from a transradial amputee using 8-channel systems for the 6 amenable motions are also classified. Statistical Analysis of Variance (ANOVA) results on classification performance of able-bodied subject divulge that the performance TD-PCA features are more significant than the AR-PCA features. Further, no significant difference in the performance of NN classifier and kNN classifier is construed with TD reduced features. Since the average classification error of kNN classifier with TD features is found to be less, kNN classifier is implemented in off-line using the TMS2407eZdsp digital signal controller to study the actuation of three low-power DC drives in the identification of intended motion with an able-bodied subject.

Modeling the Human Elbow Joint Dynamics from Surface Electromyography

Biomechanical modelling and analysis of human motion are main topics of interest for a number of disciplines, ranging from biomechanics to human movement science. There exist various experimental and theoretical techniques developed to model the biomechanics and human motor system. A classic way to characterize a system is done by perturbation analysis, through applying an external perturbation and the observation of changes in the dynamic of system. In literature, human joint dynamics has been studied mainly in relation to external perturbations. However, those perturbations interact with the natural human motor behaviour. This chapter describes an approximation for non-invasive biomechanical modelling of the elbow joint dynamics from electromyographic information. A case study presents results obtained aimed at deriving a relationship between the dynamic behaviour of the human elbow joint and Surface Electromyography (SEMG) information in postural control. A set of experiments were carried out to measure bioelectrical (SEMG) and biomechanics information from human elbow joint, during postural control (i.e. isometric contractions) and correlate them with mechanical impedance at elbow joint. Estimates of elbow impedance were obtained by applying torque perturbations to the forearm. The results demonstrate that it is possible to estimate human joint dynamics from SEMG. The obtained results can contribute to the field of human motor control and also to its application in robotics and other engineering applications through the definition, specification and characterization of properties associated with the human upper limb and strategies used by people to command it.

Design of Myocontrolled Neuroprosthesis

Recent studies suggest that the therapeutic effects of Functional Electrical Stimulation (FES) are maximized when the patterned electrical stimulation is delivered in close synchrony with the attempted voluntary movement. FES systems that modulate stimulation parameters based on the residual volitional muscle activity would assure this combination. However, the development of such a system might be not trivial, both from a hardware and a software point of view. This chapter provides an extensive overview of devices and filtering solutions proposed in the literature to estimate the residual volitional EMG signal in the presence of electrical stimulation. Different control strategies to modulate FES parameters as well as the results of the first studies involving neurological patients are also presented. This chapter provides some guidelines to help people who want to design innovative myocontrolled neuroprostheses and might favor the spread of these solutions in clinical environments.

Tongue Movement Estimation Based on Suprahyoid Muscle Activity

The motor function of the tongue often remains intact even in cases of severe movement paralysis. Therefore, tongue movements offer great potential for the design of highly efficient human-machine interfaces for alternative communication and control. This chapter introduces a novel method for tongue movement estimation based on analysis of surface electromyography (EMG) signals from the suprahyoid muscles, which usually function to open the mouth and to control the hyoid position.

Endurance Time Prediction using Electromyography

The purpose of endurance time (Tlim) prediction is to determine the exertion time of a fatiguing muscle contraction before it occurs. Tlim prediction would then allow the evaluation of muscle capacities while limiting fatigue and deleterious effects associated with exhaustive exercises. Fatigue is a progressive phenomenon which manifestations can be observed since the beginning of the exercise using electromyography (EMG). Studies have reported significant relationships between Tlim and changes in EMG signal suggesting that Tlim could be predicted from early EMG changes recorded during the first half of the fatiguing contraction. However some methodological factors can influence the reliability of the relationships between Tlim and EMG changes. The aim of this chapter is to present the methodology used to predict Tlim from early changes in EMG signal and the factors that may influence its feasibility and reliability. It will also present the possible uses and benefits of the Tlim prediction.

Study and Interpretation of Neuromuscular Patterns in Golf

This chapter reports the golf swing EMG studies using amplitude, timing parameters and approaches to neuromuscular patterns recognition through EMG. The golf swing is a dynamic multi-joint movement. During each swing phase different activation levels occur, the combination of each muscle in amplitude provides an increased club head speed for the ball to travel to the hole. The timing when the maximum peak of each muscle occurs can be an important factor to understand the injury related mechanics and to prescribe strength programs. Most muscle studies describe their maximum activation level during the forward swing and acceleration phases, providing a controlled antigravity movement and acceleration of the club. The initial contraction time corresponds to the onset that can be used to describe the organization of the neuromuscular patterns during a task. This time parameter was used in golf to relate injuries to skilled or less skilled golfers. The way to retrieve this time parameter may be reached through new approaches but no gold standard algorithm definition has been found yet. To better understand the neuromuscular patterns new algorithms based on the dynamical systems theory are now used.

Detection and Conditioning of EMG

In this chapter, the monitoring of the electrical activity of skeletal muscles is depicted. The main components of the detection and conditioning of the EMG signals is explained in the sense of the biomedical instrumentation. But, first, a brief description of EMG generation is introduced. The hardware components of the general instrumentation system used in the acquisition of EMG signal such as amplifier, filters, analog-to-digital converter are discussed in detail. Subsequently, different types of electrodes used in different EMG techniques are mentioned. Then, various EMG signals that can be detected and monitored via EMG systems are described and their clinical importance is discussed with detail. Finally, different EMG techniques used in clinical studies and their purposes are explained with detail.

The Relationship between Anthropometric Variables and Features of Electromyography Signal for Human–Computer Interface

Muscle-computer interfaces (MCIs) based on surface electromyography (EMG) pattern recognition have been developed based on two consecutive components: feature extraction and classification algorithms. Many features and classifiers are proposed and evaluated, which yield the high classification accuracy and the high number of discriminated motions under a single-session experimental condition. However, there are many limitations to use MCIs in the real-world contexts, such as the robustness over time, noise, or low-level EMG activities. Although the selection of the suitable robust features can solve such problems, EMG pattern recognition has to design and train for a particular individual user to reach high accuracy. Due to different body compositions across users, a feasibility to use anthropometric variables to calibrate EMG recognition system automatically/semi-automatically is proposed. This chapter presents the relationships between robust features extracted from actions associated with surface EMG signals and twelve related anthropometric variables. The strong and significant associations presented in this chapter could benefit a further design of the MCIs based on EMG pattern recognition.

Using in Vivo Subject-Specific Musculotendon Parameters to Investigate Voluntary Movement Changes after Stroke

Neuromusculoskeletal modeling provides insights into the muscular system which are not always obtained through experiment or observation alone. One of the major challenges in neuromusculoskeletal modeling is to accurately estimate the musculotendon parameters on a subject-specific basis. The latest medical imaging techniques such as ultrasound for the estimation of musculotendon parameters would provide an alternative method to obtain the muscle architecture parameters noninvasively. In this chapter, the feasibility of using ultrasonography to measure the musculotendon parameters of elbow muscles is validated. These parameters help to build a subject-specific EMG-driven model, which could predict the individual muscle force and elbow voluntary movement trajectory using the input of EMG signal without any trajectory fitting procedure involved. The results demonstrate the feasibility of using EMG-driven neuromusculoskeletal modeling with ultrasound-measured data for prediction of voluntary elbow movement for both unimpaired subjects and persons after stroke.

Assessing Joint Stability from Eigenvalues Obtained from Multi-Channel EMG

Electromyographic (EMG) signals have many uses. This chapter addresses the role of EMG signals to assess joint stability. Low back pain assessment and treatment interventions often involve the concepts of stability and/or joint stiffness. Using muscle activation and lumbar spine posture to calculate segmental stiffness and potential energy of the spine, eigenvalues can be linked to quantitative stability. It is reasoned that if a relationship exists between eigenvalues and individual muscles, then this approach can guide customized clinical intervention for people with defined spine instability.