Electromyography Classification during Reach-to-Grasp Motion using Manifold Learning

Abstract

Electromyography (EMG) is a simple, non-invasive, and cost-effective technology for sensing muscle activity. However, EMG is also noisy, complex, and high-dimensional. It has nevertheless been widely used in a host of human-machine-interface applications (electrical wheelchairs, virtual computer mice, prosthesis, robotic fingers, etc.) and in particular to measure reaching and grasping motions of the human hand. Here, we developd a more automated pipeline to predict object weight in a reach-and-grasp task from an open dataset relying only on EMG data. In that we shifted the focus from manual feature-engineering to automated feature-extraction by using raw (filtered) EMG signals and thus letting the algorithms select the features. We further compared intrinsic EMG features, derived from several dimensionality-reduction methods, and then ran some classification algorithms on these low-dimensional representations. We found that the Laplacian Eigenmap algorithm generally outperformed other dimensionality-reduction methods. What is more, optimal classification accuracy was achieved using a combination of Laplacian Eigenmaps (simple-minded) and k-Nearest Neighbors (88% for 3-way classification). Our results, using EMG alone, are comparable to others in the literature that used EMG and EEG together. They also demonstrate the usefulness of dimensionality reduction when classifying movement based on EMG signals and more generally the usefulness of EMG for movement classification.

An end-to-end CNN with attentional mechanism applied to raw EEG in a BCI classification task

“…In particular, it strongly enhances the scientific armamentarium used to investigate volition [3,4]. And, more specifically, decoding intention in real time would open the door to interesting experimental possibilities, such as interventions to facilitate or frustrate intentions [5][6][7], and intention-contingent stimulation [3]. Technological advances of recent decadessuch as untethered, wireless recording, machinelearning-based analysis, and real-time analysis of raw electroencephalography (EEG) signal have increased the interest in EEG based BCI approaches [8].…”

Electromyography Signal Classification Using Deep Learning

“…These signals are observed to be often very chaotic and to be strongly tied up to the intrinsic properties of the muscles themselves as they are related to their structures and their functionalities. As a result, these signals usually exhibit quite noisy, complex, and high-dimensional characteristics [3].…”