Spatially Filtered Low-Density EMG and Time-Domain Descriptors Improves Hand Movement Recognition

Modem Myoelectric Control - Is it Time to Change the Algorithmic Focus?

This paper explores the evolving landscape of Electromyogram (EMG) signal analysis, focusing on the growing prominence of deep learning (DL) algorithms for hand, wrist, and finger movement recognition. Such algorithms often come with high computational costs, potentially limiting the clinical translation on resource-limited devices and igniting more research on reduced complexity models. This prompts the question: is it time to shift the algorithmic focus in EMG pattern recognition, given the reported performance of some light-weight traditional or hybrid methods emphasizing synergy between different EMG signals? A comparative study is implemented between state-of-the-art deep learning extension for time series classification, denoted as Random Convolutional Kernel Transform (ROCKET), and simple, yet effective pattern recognition methods tailored to exploit basic forms of EMG signal synergies- Waveform Length Phasors (WLPHASOR), Root-Mean-Squared Phasor (RMSPHASOR), and the proposed novel Multi-Signal Waveform Length (MSWL). Tests are conducted on EMG data from 22 participants performing 11 hand and wrist movements using two EMG armbands (10 and 8 channels, respectively), utilizing the open-source LibEMG toolbox. Preliminary findings suggest that, while DL algorithms exhibit formidable capabilities, the performance gap with traditional EMG feature extraction methods may not be as substantial as anticipated. The observations of this study revealed no significant differences in average accuracies between ROCKET, WLPHASOR, and RMSPHASOR (87% average across participants). Furthermore, MSWL significantly enhances performance to 90%, and the combination of ROCKET+MSWL achieves 91% on average across all subjects. These findings challenge the narrative of DL dominance in EMG pattern recognition, urging a re-evaluation of the algorithmic focus and contributing valuable insights to the debate on information extraction from EMG signals.

Combined Dynamic Time Warping and Spatiotemporal Attention for Myoelectric Control

The success of pattern recognition based upper-limb prostheses control is linked to their ability to extract appropriate features from the electromyogram (EMG) signals. Traditional EMG feature extraction (FE) algorithms fail to extract spatial and inter-temporal information from the raw data, as they consider the EMG channels individually across a set of sliding windows with some degree of overlapping. To tackle these limitations, this paper presents a method that considers the spatial information of multi-channel EMG signals by utilising dynamic time warping (DTW). To satisfy temporal considerations, inspired by Long Short-Term Memory (LSTM) neural networks, our algorithm evolves the DTW feature representation across long and short-term components to capture the temporal dynamics of the EMG signal. As such the contribution of this paper is the development of a recursive spatio-temporal FE method, denoted as Recursive Temporal Warping (RTW). To investigate the performance of the proposed method, an offline EMG pattern recognition study with 53 movement classes performed by 10 subjects wearing 8 to 16 EMG channels was considered with the results compared against several conventional as well as deep learning-based models. We show that the use of the RTW can reduce classification errors significantly, paving the way for future real-time implementation.

Recurrent Fusion of Time-Domain Descriptors Improves EMG-based Hand Movement Recognition

Controlling powered prostheses with myoelectric pattern recognition (PR) provides a natural human-robot interfacing scheme for amputees who lost their limbs. Research in this direction reveals that the challenges prohibiting reliable clinical translation of myoelectric interfaces are mainly driven by the quality of the extracted features. Hence, developing accurate and reliable feature extraction techniques is of vital importance for facilitating clinical implementation of Electromyogram (EMG) PR systems. To overcome this challenge, we proposed a combination of Range Spatial Filtering (RSF) and Recurrent Fusion of Time Domain Descriptors (RFTDD) in order to improve the classifier performance and make the prosthetic hand control more appropriate for clinical applications. RSF is used to increase the number of EMG signals available for feature extraction by focusing on the spatial information between all possible logical combinations of the physical EMG channels. RFTDD is then used to capture the temporal information by applying a recurrent data fusion process on the resulting orientation-based time-domain (TD) features, with a sigmoidal function to limit the features range and overcome the vanishing amplitudes problem. The main advantages of the proposed method include 1) its potential in capturing the temporal-spatial dependencies of the EMG signals, leading to reduced classification errors, and 2) the simplicity with which the features are extracted, as any kind of simple TD features can be adopted with this method. The performance of the proposed RFTDD is then benchmarked across many well-known TD features individually and as sets to prove the power of the RFTDD method on two EMG datasets with a total of 31 subjects. Testing results revealed an approximate reduction of 12% in classification errors across all subjects when using the proposed method against traditional feature extraction methods.Clinical Relevance-Establishing significance and importance of RFTDD, with simple time-domain features, for robust and low-cost clinical applications.