Application of neural network filters for improving an electromyogram signal filtration

Electromyography is a widely applied method for measuring the electrical activity of muscles in medical diagnostics, rehabilitation, and biomechanics. However, electromyography signals often encounter challenges due to noise and interference, complicating their interpretation and analysis. This pape...

Full description

Saved in:
Bibliographic Details
Main Authors: Andrey A. Boshlyakov, Nikita A. Shilov, Maksim I. Zharkov
Format: Article
Language:English
Published: Tomsk Polytechnic University 2024-03-01
Series:Известия Томского политехнического университета: Промышленная кибернетика
Subjects:
Online Access:https://indcyb.ru/journal/article/view/48/38
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850121846461038592
author Andrey A. Boshlyakov
Nikita A. Shilov
Maksim I. Zharkov
author_facet Andrey A. Boshlyakov
Nikita A. Shilov
Maksim I. Zharkov
author_sort Andrey A. Boshlyakov
collection DOAJ
description Electromyography is a widely applied method for measuring the electrical activity of muscles in medical diagnostics, rehabilitation, and biomechanics. However, electromyography signals often encounter challenges due to noise and interference, complicating their interpretation and analysis. This paper proposes a novel approach: filtering electromyography signals using neural networks. This method effectively extracts valuable information from the signals while minimizing the impact of noise and distortion. The proposed approach involves preprocessing the signal, designing a neural network filter architecture, and training it on appropriate datasets. Experimental results demonstrate the superior efficiency of this method compared to traditional analog methods for filtering electromyography signals. The proposed signal processing method finds applications in medical analysis and rehabilitation medicine, particularly in tasks requiring precise handling of electromyography data. The paper further details the development of a data identification system aimed at determining the position of the hand of a manipulator operator and the applied force.
format Article
id doaj-art-ae9932b5c3c54cb492c181302146a595
institution OA Journals
issn 2949-5407
language English
publishDate 2024-03-01
publisher Tomsk Polytechnic University
record_format Article
series Известия Томского политехнического университета: Промышленная кибернетика
spelling doaj-art-ae9932b5c3c54cb492c181302146a5952025-08-20T02:34:59ZengTomsk Polytechnic UniversityИзвестия Томского политехнического университета: Промышленная кибернетика2949-54072024-03-0121303710.18799/29495407/2024/1/48Application of neural network filters for improving an electromyogram signal filtrationAndrey A. Boshlyakov0Nikita A. Shilov1Maksim I. Zharkov2Bauman Moscow State Technical University, Moscow, Russian FederationBauman Moscow State Technical University, Moscow, Russian FederationBauman Moscow State Technical University, Moscow, Russian FederationElectromyography is a widely applied method for measuring the electrical activity of muscles in medical diagnostics, rehabilitation, and biomechanics. However, electromyography signals often encounter challenges due to noise and interference, complicating their interpretation and analysis. This paper proposes a novel approach: filtering electromyography signals using neural networks. This method effectively extracts valuable information from the signals while minimizing the impact of noise and distortion. The proposed approach involves preprocessing the signal, designing a neural network filter architecture, and training it on appropriate datasets. Experimental results demonstrate the superior efficiency of this method compared to traditional analog methods for filtering electromyography signals. The proposed signal processing method finds applications in medical analysis and rehabilitation medicine, particularly in tasks requiring precise handling of electromyography data. The paper further details the development of a data identification system aimed at determining the position of the hand of a manipulator operator and the applied force.https://indcyb.ru/journal/article/view/48/38neural networkbiopotentialelectromyographymanipulator
spellingShingle Andrey A. Boshlyakov
Nikita A. Shilov
Maksim I. Zharkov
Application of neural network filters for improving an electromyogram signal filtration
Известия Томского политехнического университета: Промышленная кибернетика
neural network
biopotential
electromyography
manipulator
title Application of neural network filters for improving an electromyogram signal filtration
title_full Application of neural network filters for improving an electromyogram signal filtration
title_fullStr Application of neural network filters for improving an electromyogram signal filtration
title_full_unstemmed Application of neural network filters for improving an electromyogram signal filtration
title_short Application of neural network filters for improving an electromyogram signal filtration
title_sort application of neural network filters for improving an electromyogram signal filtration
topic neural network
biopotential
electromyography
manipulator
url https://indcyb.ru/journal/article/view/48/38
work_keys_str_mv AT andreyaboshlyakov applicationofneuralnetworkfiltersforimprovinganelectromyogramsignalfiltration
AT nikitaashilov applicationofneuralnetworkfiltersforimprovinganelectromyogramsignalfiltration
AT maksimizharkov applicationofneuralnetworkfiltersforimprovinganelectromyogramsignalfiltration