Locomo-Net: A Low -Complex Deep Learning Framework for sEMG-Based Hand Movement Recognition for Prosthetic Control

Abstract

Background: The enhancement in the performance of the myoelectric pattern recognition techniques based on deep learning algorithm possess computationally expensive and exhibit extensive memory behavior. Therefore, in this paper we report a deep learning framework named 'Low-Complex Movement recognition-Net' (LoCoMo-Net) built with convolution neural network (CNN) for recognition of wrist and finger flexion movements; grasping and functional movements; and force pattern from single channel surface electromyography (sEMG) recording. The network consists of a two-stage pipeline: 1) input data compression; 2) data-driven weight sharing. Methods: The proposed framework was validated on two different datasets-our own dataset (DS1) and publicly available NinaPro dataset (DS2) for 16 movements and 50 movements respectively. Further, we have prototyped the proposed LoCoMo-Net on Virtex-7 Xilinx field-programmable gate array (FPGA) platform and validated for 15 movements from DS1 to demonstrate its feasibility for real-time execution. Results: The effectiveness of the proposed LoCoMo-Net was verified by a comparative analysis against the benchmarked models using the same datasets wherein our proposed model outperformed Twin-Support Vector Machine (SVM) and existing CNN based model by an average classification accuracy of 8.5 % and 16.0 % respectively. In addition, hardware complexity analysis is done to reveal the advantages of the two-stage pipeline where approximately 27 %, 49 %, 50 %, 23 %, and 43 % savings achieved in lookup tables (LUT's), registers, memory, power consumption and computational time respectively. Conclusion: The clinical significance of such sEMG based accurate and lowcomplex movement recognition system can be favorable for the potential improvement in quality of life of an amputated persons.

High-Performance Surface Electromyography Armband Design for Gesture Recognition

“…In the pooling layer, downsampling further reduces the number of training parameters of the network, reduces the impact of pixel value changes on the convolution results, and improves the generalization ability of the network. Therefore, CNNs have better robustness and intelligence in processing images [ 21 , 22 , 23 ]. Duan Na et al [ 24 ] showed through comparative experiments that CNNs have higher gesture recognition accuracies than SVMs.…”

Hand gestures classification of sEMG signals based on BiLSTM- Metaheuristic Optimization and Hybrid U-Net-MobileNetV2 Encoder Architecture

“…Previous studies that utilized sEMG signals to analyze hand motions and gestures employed a diverse range of deep structures, including CNNs (Ding et al, 2018;Allard et al, 2016), Deep Belief Networks (DBNs) (Shim et al, 2016;Su et al, 2016), Transfer Learning (TL) (Côté-Allard et al, 2017;Du et al, 2017;Suri et al, 2018;Côté-Allard et al, 2019), Recurrent Neural Networks (RNNs) (Hu et al, 2018;Simão et al, 2019), and Adversarial Learning (AL) (Hu et al, 2019;Wei et al, 2019). Gautam et al (2020) (2020) proposed a novel three-dimensional game controlled by sEMG using a deep learning-based architecture. The main objective of their study was to develop a 3D gaming experience that could be easily manipulated with inexpensive sEMG sensors, thereby enabling individuals with disabilities to access the game.…”

Deep learning for biosignal control: insights from basic to real-time methods with recommendations

“…Both these methods contribute to improving training speed and generalization of the network, and ensure that the model converges to an optimal loss. Both batch normalization (Tayeb et al 2019, Tam et al 2020 and dropout (Gautam et al 2020, Tortora et al 2020a are often used in biosignal decoding papers, sometimes both at the same time. Other normalization techniques are possible, such as L1normalization or clipping the gradients (Zhang et al 2019a), but they have been superseded by Batch Normalization and Dropout.…”