Mechanomyogram Signals Research Articles

AI-Enabled Soft Sensing Array for Simultaneous Detection of Muscle Deformation and Mechanomyography for Metaverse Somatosensory Interaction.

Motion recognition (MR)-based somatosensory interaction technology, which interprets user movements as input instructions, presents a natural approach for promoting human-computer interaction, a critical element for advancing metaverse applications. Herein, this work introduces a non-intrusive muscle-sensing wearable device, that in conjunction with machine learning, enables motion-control-based somatosensory interaction with metaverse avatars. To facilitate MR, the proposed device simultaneously detects muscle mechanical activities, including dynamic muscle shape changes and vibrational mechanomyogram signals, utilizing a flexible 16-channel pressure sensor array (weighing ≈0.38g). Leveraging the rich information from multiple channels, a recognition accuracy of ≈96.06% is achieved by classifying ten lower-limb motions executed by ten human subjects. In addition, this work demonstrates the practical application of muscle-sensing-based somatosensory interaction, using the proposed wearable device, for enabling the real-time control of avatars in a virtual space. This study provides an alternative approach to traditional rigid inertial measurement units and electromyography-based methods for achieving accurate human motion capture, which can further broaden the applications of motion-interactive wearable devices for the coming metaverse age.

Gesture Recognition through Mechanomyogram Signals: An Adaptive Framework for Arm Posture Variability.

In hand gesture recognition, classifying gestures across multiple arm postures is challenging due to the dynamic nature of muscle fibers and the need to capture muscle activity through electrical connections with the skin. This paper presents a gesture recognition architecture addressing the arm posture challenges using an unsupervised domain adaptation technique and a wearable mechanomyogram (MMG) device that does not require electrical contact with the skin. To deal with the transient characteristics of muscle activities caused by changing arm posture, Continuous Wavelet Transform (CWT) combined with Domain-Adversarial Convolutional Neural Networks (DACNN) were used to extract MMG features and classify hand gestures. DACNN was compared with supervised trained classifiers and shown to achieve consistent improvement in classification accuracies over multiple arm postures. With less than 5 minutes of setup time to record 20 examples per gesture in each arm posture, the developed method achieved an average prediction accuracy of 87.43% for classifying 5 hand gestures in the same arm posture and 64.29% across 10 different arm postures. When further expanding the MMG segmentation window from 200 ms to 600 ms to extract greater discriminatory information at the expense of longer response time, the intraposture and inter-posture accuracies increased to 92.32% and 71.75%. The findings demonstrate the capability of the proposed method to improve generalization throughout dynamic changes caused by arm postures during non-laboratory usages and the potential of MMG to be an alternative sensor with comparable performance to the widely used electromyogram (EMG) gesture recognition systems.

An Elbow Motion Classification Approach Based on the MT System Using Mechanomyogram Signals

The objective of this paper is to propose a mechanomyogram (MMG)-based motion classification system comprised of a muscle-activity onset detector and a motion classifier. The detector identifies mu...

Automatic detection of muscle activity from mechanomyogram signals: a comparison of amplitude and wavelet-based methods

Knowledge of muscle activity timing is critical to many clinical applications, such as the assessment of muscle coordination and the prescription of muscle-activated switches for individuals with disabilities. In this study, we introduce a continuous wavelet transform (CWT) algorithm for the detection of muscle activity via mechanomyogram (MMG) signals. CWT coefficients of the MMG signal were compared to scale-specific thresholds derived from the baseline signal to estimate the timing of muscle activity. Test signals were recorded from the flexor carpi radialis muscles of 15 able-bodied participants as they squeezed and released a hand dynamometer. Using the dynamometer signal as a reference, the proposed CWT detection algorithm was compared against a global-threshold CWT detector as well as amplitude-based event detection for sensitivity and specificity to voluntary contractions. The scale-specific CWT-based algorithm exhibited superior detection performance over the other detectors. CWT detection also showed good muscle selectivity during hand movement, particularly when a given muscle was the primary facilitator of the contraction. This may suggest that, during contraction, the compound MMG signal has a recurring morphological pattern that is not prevalent in the baseline signal. The ability of CWT analysis to be implemented in real time makes it a candidate for muscle-activity detection in clinical applications.

Stationarity distributions of mechanomyogram signals from isometric contractions of extrinsic hand muscles during functional grasping

This study investigates the stationarity of steady state mechanomyogram signals for the purpose of determining appropriate features for signal classification. Mechanomyography is the superficial recording of low frequency vibrations detected over contracting muscles. Steady state mechanomyogram signals, recorded at the belly of the extensor digitorum, flexor digitorum superficialis and flexor pollicis longus muscles during functional grasps were tested for weak stationarity. Twenty percent of the contractions were found to be non-stationary, indicating that time frequency methods may be appropriate for automatic pattern recognition of functional grasp from the mechanomyogram. The distribution of the stationary test statistic was dependent on the type of muscle contractions, suggesting that the test statistic itself might be a discriminating feature for mechanomyogram pattern recognition in applications such as multifunction prosthetic control. Since the major known source of non-stationarity was decreasing variance, it is suggested that shifts in the distribution of the test statistic may indicate the time course of relative muscle contributions to functional grasp.

Relationship between mechanomyogram signals and changes in force of human forefinger flexor muscles during voluntary contraction.

In previous studies on mechanomyogram (MMG) signals no analysis of these signals accompanying force generation has been performed. Therefore, we have recorded MMG signals (previously referred to as muscle sound or acoustomyographic signals) during voluntary contractions of forefinger flexor muscles in 31 young subjects. These subjects made contractions to produce force records of triangular or trapeziform shape. The peak target force amounted to 10, 20 or 40 N which represented less than 40% of maximal voluntary contraction. The MMG signals during the transient phases of force generation at three different rates were analysed. The MMG intensity level calculated for MMG records and the peak-to-peak amplitude of MMG signals correlated with both the velocity of force increase and the contraction force. The occurrence of the strongest MMG signals corresponded to changes in contractile force. Therefore, it is suggested that measurements of these parameters could be a useful tool in studies of changes in contractile force.