Lower-Limb Motion-Based Ankle-Foot Movement Classification Using 2D-CNN

作     者：Narathip Chaobankoh Tallit Jumphoo Monthippa Uthansakul Khomdet Phapatanaburi Bura Sindthupakorn Supakit Rooppakhun Peerapong Uthansakul 

作者机构：School of Telecommunication EngineeringInstitute of EngineeringSuranaree University of TechnologyNakhon Ratchasima30000Thailand Department of Telecommunications EngineeringFaculty of Engineering and TechnologyRajamangala University of Technology Isan(RMUTI)Nakhon Ratchasima30000Thailand Orthopedic Department School of MedicineSuranaree University of TechnologyNakhon Ratchasima30000Thailand School of Mechanical EngineeringInstitute of EngineeringSuranaree University of TechnologyNakhon Ratchasima30000Thailand 

出 版 物：《Computers, Materials & Continua》 (计算机、材料和连续体（英文）)

年 卷 期：2022年第73卷第10期

页      面：1269-1282页

核心收录：

学科分类：1002[医学-临床医学] 10[医学] 

基　　金：This work was supported by Suranaree University of Technology(SUT) Thailand Science Research and Innovation(TSRI) and National Science Research and Innovation Fund(NSRF)(NRIIS no.42852) 

主　　题：Electromyography neural network tibialis anterior muscle gastrocnemius muscle convolution neural network spectrogram lower limb 

摘      要：Recently,the Muscle-Computer Interface(MCI)has been extensively popular for employing Electromyography(EMG)signals to help the development of various assistive ***,few studies have focused on ankle foot movement classification considering EMG signals at limb *** work proposes a new framework considering two EMG signals at a lower-limb position to classify the ankle movement characteristics based on normal walking *** this purpose,we introduce a human anklefoot movement classification method using a two-dimensional-convolutional neural network(2D-CNN)with low-cost EMG sensors based on lowerlimb *** time-domain signals of EMG obtained from two sensors belonging to Dorsiflexion,Neutral-position,and Plantarflexion are firstly converted into time-frequency spectrograms by short-time Fourier ***,the spectrograms of the three ankle-foot movement types are used as input to the 2D-CNN such that the EMG foot movement types are finally *** the evaluation phase,the proposed method is investigated using the healthy volunteer for 5-fold cross-validation,and the accuracy is used as a standard *** results demonstrate that our approach provides an average accuracy of 99.34%.This exhibits the usefulness of 2D-CNN with low-cost EMG sensors in terms of ankle-foot movement classification at limb position,which offers feasibility for ***,the obtained EMG signal is not directly considered at the ankle position.