A Strategy for Identifying Locomotion Modes Using Surface Electromyography

Abstract

This study investigated the use of surface electromyography (EMG) combined with pattern recognition (PR) to identify user locomotion modes. Due to the nonstationary characteristics of leg EMG signals during locomotion, a new phase-dependent EMG PR strategy was proposed for classifying the user's locomotion modes. The variables of the system were studied for accurate classification and timely system response. The developed PR system was tested on EMG data collected from eight able-bodied subjects and two subjects with long transfemoral (TF) amputations while they were walking on different terrains or paths. The results showed reliable classification for the seven tested modes. For eight able-bodied subjects, the average classification errors in the four defined phases using ten electrodes located over the muscles above the knee (simulating EMG from the residual limb of a TF amputee) were 12.4% ± 5.0%, 6.0% ± 4.7%, 7.5% ± 5.1%, and 5.2% ± 3.7%, respectively. Comparable results were also observed in our pilot study on the subjects with TF amputations. The outcome of this investigation could promote the future design of neural-controlled artificial legs.

Simultaneous Bayesian Recognition of Locomotion and Gait Phases With Wearable Sensors

“…An adaptive algorithm, based on decision trees and four sensors attached to the human body, was implemented for recognition of daily activities such as walking, standing and sitting with an accuracy of 99% [16]. Fusion of a linear discriminant analysis (LDA) method and a two-layered artificial neural network (ANN), was used for identification of locomotion modes with twelve surface EMG signals [17]. LDA and ANN methods have also been used with time-domain and frequency-domain features from nine EMG signals for intent recognition [18], [19].…”

A combined Adaptive Neuro-Fuzzy and Bayesian strategy for recognition and prediction of gait events using wearable sensors

“…Specifically, large progress has been observed in wearable sensors -for instance, lightweight and fast inertial measurement units (IMUs) and soft kinematic sensors [6], [7], [8]. In contrast, the deployment of computational methods that permit to perform fast and accurate human motion analysis, recognition of walking activities and prediction of gait events are still under development [9], [10], [11]. In this work, we present a novel strategy for recognition and prediction of gait events, that combines two computational intelligence approaches: Adaptive Neuro-Fuzzy and Bayesian methods (Figure 1).…”

Towards Intelligent Lower Limb Prostheses with Activity Recognition

“…High classification accuracies can be achieved by this approach, however; it requires an extensive collection of data for training the classifier [4]. The main challenge in the powered devices is the lack of direct control by amputees [5]. Therefore, the need to control the prostheses intuitively has brought the ideas of using surface electromyography (sEMG), mechanical sensors or a fusion-based control.…”

“…Surface EMG (sEMG) electrodes have been used to record muscle activities signals from amputees wearing passive prostheses and powered prostheses [6]. Several studies investigated EMG PR to identify the user intent in different activities [5,7,8] for smooth, intuitive and natural control of prostheses. A number of studies have reported the use of mechanical sensors (inertial measurement units (IMUs)), load sensors and pressure-sensitive insoles) for lower limb activity recognition [6,9,10].…”