Generalizing Upper Limb Force Modeling With Transfer Learning: A Multimodal Approach Using EMG and IMU for New Users and Conditions.

In the field of EMG-based force modeling, the ability to generalize models across individuals could play a significant role in its adoption across a range of applications, including assistive devices, robotic and rehabilitation devices. However, current studies have predominately focused on intra-subject modeling, largely neglecting the burden of end-user data acquisition. In this work, we propose the use of transfer learning (TL) to generalize force modeling to a new user by first establishing a baseline model trained using other users' data, and then adapting to the end-user using a small amount of new data (only 10% , 20% , and 40% of the new user data). Using a deep multimodal convolutional neural network, consisting of two CNN models, one with high-density (HD) EMG and one with motion data recorded by an Inertial Measurement Unit (IMU), our proposed TL technique significantly improved force modeling compared to leave-one-subject-out (LOSO) and even intra-subject scenarios. The TL approach increased the average R squared values of the force modeling task by 60.81%, 190.53%, and 199.79% compared to the LOSO case, and by 13.4%, 36.88%, and 45.51% compared to the intra-subject case for isotonic, isokinetic and dynamic conditions, respectively. These results show that it is possible to adapt to a new user with minimal data while improving performance significantly compared to the intra-subject scenario. We also show that TL can be used to generalize on a new experimental condition for a new user.

Tutorial. Surface electromyogram (sEMG) amplitude estimation: Best practices.

This tutorial intends to provide insight, instructions and "best practices" for those who are novices-including clinicians, engineers and non-engineers-in extracting electromyogram (EMG) amplitude from the bipolar surface EMG (sEMG) signal of voluntary contractions. A brief discussion of sEMG amplitude extraction from high density sEMG (HDsEMG) arrays and feature extraction from electrically elicited contractions is also provided. This tutorial attempts to present its main concepts in a straightforward manner that is accessible to novices in the field not possessing a wide range of technical background (if any) in this area. Surface EMG amplitude, also referred to as the sEMG envelope [often implemented as root mean square (RMS) sEMG or average rectified value (ARV) sEMG], quantifies the voltage variation of the sEMG signal and is grossly related to the overall neural excitation of the muscle and to peripheral parameters. The tutorial briefly reviews the physiological origin of the voluntary sEMG signal and sEMG recording, including electrode configurations, sEMG signal transduction, electronic conditioning and conversion by an analog-to-digital converter. These topics have been covered in greater detail in prior tutorials in this series. In depth descriptions of state-of-the-art methods for computing sEMG amplitude are then provided, including guidance on signal pre-conditioning, absolute value vs. square-law detection, selection of appropriate sEMG amplitude smoothing filters and attenuation of measurement noise. The tutorial provides a detailed list of best practices for sEMG amplitude estimation.

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

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 R2 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.

Convolutional Neural Network Approach for Elbow Torque Estimation during Quasi-dynamic and Dynamic Contractions.

Accurate torque estimation during dynamic conditions is challenging, yet an important problem for many applications such as robotics, prosthesis control, and clinical diagnostics. Our objective is to accurately estimate the torque generated at the elbow during flexion and extension, under quasi-dynamic and dynamic conditions. High-density surface electromyogram (HD-EMG) signals, acquired from the long head and short head of biceps brachii, brachioradialis, and triceps brachii of five participants are used to estimate the torque generated at the elbow, using a convolutional neural network (CNN). We hypothesise that incorporating the mechanical information recorded by the biodex machine, i.e., position and velocity, can improve the model performance. To investigate the effects of the added data modalities on the model accuracy, models are constructed that combine EMG and position, as well as EMG with both position and velocity. R2 values are improved by 2.35%, 37.50%, and 16.67%, when position and EMG are used as inputs to the CNN models, for isotonic, isokinetic, and dynamic cases, respectively compared to using only EMG. The model performances improves further by 2.29%, 12.12%, and 20.50% for isotonic, isokinetic, and dynamic conditions, when velocity is added with the EMG and position data.

Automated Channel Selection in High-Density sEMG for Improved Force Estimation.

Accurate and real-time estimation of force from surface electromyogram (EMG) signals enables a variety of applications. We developed and validated new approaches for selecting subsets of high-density (HD) EMG channels for improved and lower-dimensionality force estimation. First, a large dataset was recorded from a number of participants performing isometric contractions in different postures, while simultaneously recording HD-EMG channels and ground-truth force. The EMG signals were acquired from three linear surface electrode arrays, each with eight monopolar channels, and were placed on the long head and short head of the biceps brachii and brachioradialis. After data collection and pre-processing, fast orthogonal search (FOS) was employed for force estimation. To select a subset of channels, principal component analysis (PCA) in the frequency domain and a novel index called the power-correlation ratio (PCR), which maximizes the spectral power while minimizing similarity to other channels, were used. These approaches were compared to channel selection using time-domain PCA. We selected one, two, and three channels per muscle from the original seven differential channels to reduce the redundancy and correlation in the dataset. In the best case, we achieved an approximate improvement of 30% for force estimation while reducing the dimensionality by 57% for a subset of three channels.

PCA-Based Channel Selection in High-Density EMG for Improving Force Estimation.

In this paper, a method for selecting channels to improve estimated force using fast orthogonal search (FOS) has been proposed. Surface electromyogram (sEMG) signals acquired from linear surface electrode arrays, placed on the long head and short head of biceps brachii, and brachioradialis during isometric contractions are used to estimate force induced at the wrist using the FOS algorithm. The method utilizes principle component analysis (PCA) in the frequency domain to select the channels with the highest contribution to the first principal component (PC). Our analysis demonstrates that our proposed method is capable of reducing the dimensionality of the data (the number of channels was reduced from 21 to 9) while improving the accuracy of the estimated force.

An Investigation of Dimensionality Reduction Techniques for EMG-based Force Estimation.

In this paper, extracted features in time and frequency domain, from high-density surface electromyogram (HD-sEMG) signals acquired from the long head and short head of biceps brachii, and brachioradialis during isometric elbow flexion are used to estimate force induced at the wrist using an artificial neural network (ANN). Different hidden layer sizes were considered to investigate its effect on the model accuracy. Also, we applied two dimensionality reduction techniques, principal component analysis (PCA) and t-distributed stochastic neighbor embedding (t-SNE), on the feature set and investigated their effects on force estimation accuracy.