A Neural Network Based on the Johnson <italic>S</italic><sub>U</sub> Translation System and Related Application to Electromyogram Classification

Electromyogram (EMG) classification is a key technique in EMG-based control systems. Existing EMG classification methods, which do not consider EMG features that have distribution with skewness and kurtosis, have limitations such as the requirement to tune hyperparameters. In this paper, we propose...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Hideaki Hayashi, Taro Shibanoki, Toshio Tsuji
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Biomedical signal processing electromyography Johnson distribution neural networks pattern recognition Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/4b9066e93ed74ac1a10ba74c28a8b2ee
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:4b9066e93ed74ac1a10ba74c28a8b2ee
record_format	dspace
spelling	oai:doaj.org-article:4b9066e93ed74ac1a10ba74c28a8b2ee2021-11-24T00:01:45ZA Neural Network Based on the Johnson <italic>S</italic><sub>U</sub> Translation System and Related Application to Electromyogram Classification2169-353610.1109/ACCESS.2021.3126348https://doaj.org/article/4b9066e93ed74ac1a10ba74c28a8b2ee2021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9606727/https://doaj.org/toc/2169-3536Electromyogram (EMG) classification is a key technique in EMG-based control systems. Existing EMG classification methods, which do not consider EMG features that have distribution with skewness and kurtosis, have limitations such as the requirement to tune hyperparameters. In this paper, we propose a neural network based on the Johnson <inline-formula> <tex-math notation="LaTeX">$S_{\mathrm {U}}$ </tex-math></inline-formula> translation system that is capable of representing distributions with skewness and kurtosis. The Johnson system is a normalizing translation that transforms non-normal distribution data into normal distribution data, thereby enabling the representation of a wide range of distributions. In this study, a discriminative model based on the multivariate Johnson <inline-formula> <tex-math notation="LaTeX">$S_{\mathrm {U}}$ </tex-math></inline-formula> translation system is transformed into a linear combination of coefficients and input vectors using log-linearization; then, it is incorporated into a neural network structure. This allows the calculation of the posterior probability of each class given the input vectors and the determination of model parameters as weight coefficients of the network. The uniqueness of convergence of the network learning is theoretically guaranteed. In the experiments, the suitability of the proposed network for distributions including skewness and kurtosis was evaluated using artificially generated data. Its applicability to real biological data was also evaluated via EMG classification experiments. The results showed that the proposed network achieved high classification performance (e.g., 99.973% accuracy using Khushaba’s dataset) without the need for hyperparameter optimization.Hideaki HayashiTaro ShibanokiToshio TsujiIEEEarticleBiomedical signal processingelectromyographyJohnson distributionneural networkspattern recognitionElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 154304-154317 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Biomedical signal processing electromyography Johnson distribution neural networks pattern recognition Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Biomedical signal processing electromyography Johnson distribution neural networks pattern recognition Electrical engineering. Electronics. Nuclear engineering TK1-9971 Hideaki Hayashi Taro Shibanoki Toshio Tsuji A Neural Network Based on the Johnson <italic>S</italic><sub>U</sub> Translation System and Related Application to Electromyogram Classification
description	Electromyogram (EMG) classification is a key technique in EMG-based control systems. Existing EMG classification methods, which do not consider EMG features that have distribution with skewness and kurtosis, have limitations such as the requirement to tune hyperparameters. In this paper, we propose a neural network based on the Johnson <inline-formula> <tex-math notation="LaTeX">$S_{\mathrm {U}}$ </tex-math></inline-formula> translation system that is capable of representing distributions with skewness and kurtosis. The Johnson system is a normalizing translation that transforms non-normal distribution data into normal distribution data, thereby enabling the representation of a wide range of distributions. In this study, a discriminative model based on the multivariate Johnson <inline-formula> <tex-math notation="LaTeX">$S_{\mathrm {U}}$ </tex-math></inline-formula> translation system is transformed into a linear combination of coefficients and input vectors using log-linearization; then, it is incorporated into a neural network structure. This allows the calculation of the posterior probability of each class given the input vectors and the determination of model parameters as weight coefficients of the network. The uniqueness of convergence of the network learning is theoretically guaranteed. In the experiments, the suitability of the proposed network for distributions including skewness and kurtosis was evaluated using artificially generated data. Its applicability to real biological data was also evaluated via EMG classification experiments. The results showed that the proposed network achieved high classification performance (e.g., 99.973% accuracy using Khushaba’s dataset) without the need for hyperparameter optimization.
format	article
author	Hideaki Hayashi Taro Shibanoki Toshio Tsuji
author_facet	Hideaki Hayashi Taro Shibanoki Toshio Tsuji
author_sort	Hideaki Hayashi
title	A Neural Network Based on the Johnson <italic>S</italic><sub>U</sub> Translation System and Related Application to Electromyogram Classification
title_short	A Neural Network Based on the Johnson <italic>S</italic><sub>U</sub> Translation System and Related Application to Electromyogram Classification
title_full	A Neural Network Based on the Johnson <italic>S</italic><sub>U</sub> Translation System and Related Application to Electromyogram Classification
title_fullStr	A Neural Network Based on the Johnson <italic>S</italic><sub>U</sub> Translation System and Related Application to Electromyogram Classification
title_full_unstemmed	A Neural Network Based on the Johnson <italic>S</italic><sub>U</sub> Translation System and Related Application to Electromyogram Classification
title_sort	neural network based on the johnson <italic>s</italic><sub>u</sub> translation system and related application to electromyogram classification
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/4b9066e93ed74ac1a10ba74c28a8b2ee
work_keys_str_mv	AT hideakihayashi aneuralnetworkbasedonthejohnsonitalicsitalicsubusubtranslationsystemandrelatedapplicationtoelectromyogramclassification AT taroshibanoki aneuralnetworkbasedonthejohnsonitalicsitalicsubusubtranslationsystemandrelatedapplicationtoelectromyogramclassification AT toshiotsuji aneuralnetworkbasedonthejohnsonitalicsitalicsubusubtranslationsystemandrelatedapplicationtoelectromyogramclassification AT hideakihayashi neuralnetworkbasedonthejohnsonitalicsitalicsubusubtranslationsystemandrelatedapplicationtoelectromyogramclassification AT taroshibanoki neuralnetworkbasedonthejohnsonitalicsitalicsubusubtranslationsystemandrelatedapplicationtoelectromyogramclassification AT toshiotsuji neuralnetworkbasedonthejohnsonitalicsitalicsubusubtranslationsystemandrelatedapplicationtoelectromyogramclassification
_version_	1718416090790363136

A Neural Network Based on the Johnson <italic>S</italic><sub>U</sub> Translation System and Related Application to Electromyogram Classification

Ejemplares similares