Deep Multi-Scale Fusion of Convolutional Neural Networks for EMG-Based Movement Estimation

Abstract

EMG-based motion estimation is required for applications such as myoelectric control, where the simultaneous estimation of kinematic information, namely joint angle and velocity, is challenging and critical. We propose a novel method for accurately modelling the generated joint angle and velocity simultaneously under isotonic, isokinetic (quasi-dynamic), and fully dynamic conditions. Our solution uses two streams of CNN, called TS-CNN to learn informative features from raw EMG data using different scales and estimate the generated motion during elbow flexion and extension. The experimental results show the robustness of our approach in comparison to conventional CNN as well as some methods used in the literature. The best obtained R 2 values, are 0.81±0.06, 0.71±0.06, and 0.80±0.13 for joint angle estimation and 0.78±0.05, 0.79±0.07, and 0.71±0.13 for the velocity estimation, during isotonic, isokinetic, and dynamic contractions, respectively. Additionally, our results indicate that the experimental condition can have an impact on the model's performance for motion prediction. EMG-based velocity estimation obtains higher performance than joint angle estimation under isokinetic conditions. Under dynamic conditions, joint angle estimation is more accurate than velocity estimation, and there is no difference between joint angle and velocity estimation in the isotonic case.

E2CNN: An Efficient Concatenated CNN for Classification of Surface EMG Extracted From Upper Limb

“…Using a method that is based on the Short Time Fourier Transform (STFT) representation of EMG data and convolutional neural networks (CNN), Sengur et al [17] were able to attain an accuracy of 96.69%. In [18], the authors have suggested two streams-CNN (TS-CNN) to acquire significant features from raw EMG data using multiple scales, as well as estimate the motion that is created during elbow flexion and extension. The results that were obtained were 81%±0.06, 71%±0.06, and 80%±0.13 for the estimate of the joint angle, and 78%±0.05, 79%±0.07, and 71%±0.13 for the estimation of the velocity, during isotonic contractions, isokinetic contractions, and dynamic contractions, respectively.…”

Continuous Lower Limb Multi‐Intent Prediction for Electromyography‐Driven Intrinsic and Extrinsic Control

“…Flexion-extension (FE) angles are recognized based on EMG features extracted using deep belief networks and convolutional neural networks. [12,13] Some researchers mapped EMG to the joint torque and the performances of neural networks, polynomial regression, and linear regression were compared. [14,15] These methods also lack information on robotic control to improve the control switches.…”

Multimodal Estimation of End Point Force During Quasi-dynamic and Dynamic Muscle Contractions Using Deep Learning

“…Since the demonstration of the relationship between muscle electrical activity and generated force in humans [1], the surface electromyogram (EMG) has been widely used as a non-invasive estimator of the generated force and joint torque. EMG signals can be used to estimate the neural command for muscle contraction and force generation in motor control [2], and for designing control algorithms based on the user's intention for robotic arm [3], [4], powered exoskeleton [5], and prosthesis control [6]- [8] applications. In ergonomics [9] and clinical biomechanics [10] EMG can be used to better estimate joint loading and muscle tension to prevent musculoskeletal injuries.…”