A multi-label deep residual shrinkage network for high-density surface electromyography decomposition in real-time

Background The swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time control in neural interfaces. However, the existing sEMG decomposition methods, including blind source separation (BSS) and deep learning, have...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Journal of neuroengineering and rehabilitation Ročník 22; číslo 1; s. 106 - 19
Hlavní autori: Ma, Jinting, Wang, Lifen, Wu, Renxiang, Zhang, Naiwen, Wei, Jing, Li, Jianjun, Li, Qiuyuan, Tan, Lihai, Li, Guanglin, Jiang, Naifu, Dan, Guo
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: London BioMed Central 08.05.2025
BioMed Central Ltd
BMC
Predmet:
ISSN:1743-0003, 1743-0003
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Background The swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time control in neural interfaces. However, the existing sEMG decomposition methods, including blind source separation (BSS) and deep learning, have not yet achieved satisfactory performance, due to high latency or low accuracy. Methods This study introduces a novel real-time high-density sEMG (HD-sEMG) decomposition algorithm named ML-DRSNet, which combines multi-label learning with a deep residual shrinkage network (DRSNet) to improve accuracy and reduce latency. ML-DRSNet was evaluated on a public sEMG dataset and the corresponding MUSTs extracted via the convolutional BSS algorithm. An improved multi-label deep convolutional neural network (ML-DCNN) was also evaluated and compared against a conventional multi-task DCNN (MT-DCNN). These networks were trained and tested on various window sizes and step sizes. Results With the shortest window size (20 data points) and step size (10 data points), ML-DRSNet significantly outperformed both ML-DCNN (0.86 ± 0.18 vs. 0.71 ± 0.24, P  < 0.001) and MT-DCNN (0.86 ± 0.18 vs. 0.66 ± 0.16, P  < 0.001) in decomposition precision. Moreover, ML-DRSNet demonstrated a notably lower latency (15.15 ms) compared to ML-DCNN (69.36 ms) and MT-DCNN (76.96 ms), both of which demonstrated reduced latency relative to BSS-based decomposition methods. Conclusions The proposed ML-DRSNet and the improved ML-DCNN algorithms substantially enhance both the accuracy and real-time performance in decomposing MUSTs, establishing a technical foundation for neuro-information-driven motor intention recognition and disease assessment.
AbstractList Abstract Background The swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time control in neural interfaces. However, the existing sEMG decomposition methods, including blind source separation (BSS) and deep learning, have not yet achieved satisfactory performance, due to high latency or low accuracy. Methods This study introduces a novel real-time high-density sEMG (HD-sEMG) decomposition algorithm named ML-DRSNet, which combines multi-label learning with a deep residual shrinkage network (DRSNet) to improve accuracy and reduce latency. ML-DRSNet was evaluated on a public sEMG dataset and the corresponding MUSTs extracted via the convolutional BSS algorithm. An improved multi-label deep convolutional neural network (ML-DCNN) was also evaluated and compared against a conventional multi-task DCNN (MT-DCNN). These networks were trained and tested on various window sizes and step sizes. Results With the shortest window size (20 data points) and step size (10 data points), ML-DRSNet significantly outperformed both ML-DCNN (0.86 ± 0.18 vs. 0.71 ± 0.24, P < 0.001) and MT-DCNN (0.86 ± 0.18 vs. 0.66 ± 0.16, P < 0.001) in decomposition precision. Moreover, ML-DRSNet demonstrated a notably lower latency (15.15 ms) compared to ML-DCNN (69.36 ms) and MT-DCNN (76.96 ms), both of which demonstrated reduced latency relative to BSS-based decomposition methods. Conclusions The proposed ML-DRSNet and the improved ML-DCNN algorithms substantially enhance both the accuracy and real-time performance in decomposing MUSTs, establishing a technical foundation for neuro-information-driven motor intention recognition and disease assessment.
The swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time control in neural interfaces. However, the existing sEMG decomposition methods, including blind source separation (BSS) and deep learning, have not yet achieved satisfactory performance, due to high latency or low accuracy. This study introduces a novel real-time high-density sEMG (HD-sEMG) decomposition algorithm named ML-DRSNet, which combines multi-label learning with a deep residual shrinkage network (DRSNet) to improve accuracy and reduce latency. ML-DRSNet was evaluated on a public sEMG dataset and the corresponding MUSTs extracted via the convolutional BSS algorithm. An improved multi-label deep convolutional neural network (ML-DCNN) was also evaluated and compared against a conventional multi-task DCNN (MT-DCNN). These networks were trained and tested on various window sizes and step sizes. With the shortest window size (20 data points) and step size (10 data points), ML-DRSNet significantly outperformed both ML-DCNN (0.86 ± 0.18 vs. 0.71 ± 0.24, P < 0.001) and MT-DCNN (0.86 ± 0.18 vs. 0.66 ± 0.16, P < 0.001) in decomposition precision. Moreover, ML-DRSNet demonstrated a notably lower latency (15.15 ms) compared to ML-DCNN (69.36 ms) and MT-DCNN (76.96 ms), both of which demonstrated reduced latency relative to BSS-based decomposition methods. The proposed ML-DRSNet and the improved ML-DCNN algorithms substantially enhance both the accuracy and real-time performance in decomposing MUSTs, establishing a technical foundation for neuro-information-driven motor intention recognition and disease assessment.
Background The swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time control in neural interfaces. However, the existing sEMG decomposition methods, including blind source separation (BSS) and deep learning, have not yet achieved satisfactory performance, due to high latency or low accuracy. Methods This study introduces a novel real-time high-density sEMG (HD-sEMG) decomposition algorithm named ML-DRSNet, which combines multi-label learning with a deep residual shrinkage network (DRSNet) to improve accuracy and reduce latency. ML-DRSNet was evaluated on a public sEMG dataset and the corresponding MUSTs extracted via the convolutional BSS algorithm. An improved multi-label deep convolutional neural network (ML-DCNN) was also evaluated and compared against a conventional multi-task DCNN (MT-DCNN). These networks were trained and tested on various window sizes and step sizes. Results With the shortest window size (20 data points) and step size (10 data points), ML-DRSNet significantly outperformed both ML-DCNN (0.86 ± 0.18 vs. 0.71 ± 0.24, P  < 0.001) and MT-DCNN (0.86 ± 0.18 vs. 0.66 ± 0.16, P  < 0.001) in decomposition precision. Moreover, ML-DRSNet demonstrated a notably lower latency (15.15 ms) compared to ML-DCNN (69.36 ms) and MT-DCNN (76.96 ms), both of which demonstrated reduced latency relative to BSS-based decomposition methods. Conclusions The proposed ML-DRSNet and the improved ML-DCNN algorithms substantially enhance both the accuracy and real-time performance in decomposing MUSTs, establishing a technical foundation for neuro-information-driven motor intention recognition and disease assessment.
Background The swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time control in neural interfaces. However, the existing sEMG decomposition methods, including blind source separation (BSS) and deep learning, have not yet achieved satisfactory performance, due to high latency or low accuracy. Methods This study introduces a novel real-time high-density sEMG (HD-sEMG) decomposition algorithm named ML-DRSNet, which combines multi-label learning with a deep residual shrinkage network (DRSNet) to improve accuracy and reduce latency. ML-DRSNet was evaluated on a public sEMG dataset and the corresponding MUSTs extracted via the convolutional BSS algorithm. An improved multi-label deep convolutional neural network (ML-DCNN) was also evaluated and compared against a conventional multi-task DCNN (MT-DCNN). These networks were trained and tested on various window sizes and step sizes. Results With the shortest window size (20 data points) and step size (10 data points), ML-DRSNet significantly outperformed both ML-DCNN (0.86 ± 0.18 vs. 0.71 ± 0.24, P < 0.001) and MT-DCNN (0.86 ± 0.18 vs. 0.66 ± 0.16, P < 0.001) in decomposition precision. Moreover, ML-DRSNet demonstrated a notably lower latency (15.15 ms) compared to ML-DCNN (69.36 ms) and MT-DCNN (76.96 ms), both of which demonstrated reduced latency relative to BSS-based decomposition methods. Conclusions The proposed ML-DRSNet and the improved ML-DCNN algorithms substantially enhance both the accuracy and real-time performance in decomposing MUSTs, establishing a technical foundation for neuro-information-driven motor intention recognition and disease assessment. Keywords: Real-time decomposition, Motor unit spike trains, High-density surface electromyography, Multi-label learning, Deep residual shrinkage network
The swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time control in neural interfaces. However, the existing sEMG decomposition methods, including blind source separation (BSS) and deep learning, have not yet achieved satisfactory performance, due to high latency or low accuracy.BACKGROUNDThe swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time control in neural interfaces. However, the existing sEMG decomposition methods, including blind source separation (BSS) and deep learning, have not yet achieved satisfactory performance, due to high latency or low accuracy.This study introduces a novel real-time high-density sEMG (HD-sEMG) decomposition algorithm named ML-DRSNet, which combines multi-label learning with a deep residual shrinkage network (DRSNet) to improve accuracy and reduce latency. ML-DRSNet was evaluated on a public sEMG dataset and the corresponding MUSTs extracted via the convolutional BSS algorithm. An improved multi-label deep convolutional neural network (ML-DCNN) was also evaluated and compared against a conventional multi-task DCNN (MT-DCNN). These networks were trained and tested on various window sizes and step sizes.METHODSThis study introduces a novel real-time high-density sEMG (HD-sEMG) decomposition algorithm named ML-DRSNet, which combines multi-label learning with a deep residual shrinkage network (DRSNet) to improve accuracy and reduce latency. ML-DRSNet was evaluated on a public sEMG dataset and the corresponding MUSTs extracted via the convolutional BSS algorithm. An improved multi-label deep convolutional neural network (ML-DCNN) was also evaluated and compared against a conventional multi-task DCNN (MT-DCNN). These networks were trained and tested on various window sizes and step sizes.With the shortest window size (20 data points) and step size (10 data points), ML-DRSNet significantly outperformed both ML-DCNN (0.86 ± 0.18 vs. 0.71 ± 0.24, P < 0.001) and MT-DCNN (0.86 ± 0.18 vs. 0.66 ± 0.16, P < 0.001) in decomposition precision. Moreover, ML-DRSNet demonstrated a notably lower latency (15.15 ms) compared to ML-DCNN (69.36 ms) and MT-DCNN (76.96 ms), both of which demonstrated reduced latency relative to BSS-based decomposition methods.RESULTSWith the shortest window size (20 data points) and step size (10 data points), ML-DRSNet significantly outperformed both ML-DCNN (0.86 ± 0.18 vs. 0.71 ± 0.24, P < 0.001) and MT-DCNN (0.86 ± 0.18 vs. 0.66 ± 0.16, P < 0.001) in decomposition precision. Moreover, ML-DRSNet demonstrated a notably lower latency (15.15 ms) compared to ML-DCNN (69.36 ms) and MT-DCNN (76.96 ms), both of which demonstrated reduced latency relative to BSS-based decomposition methods.The proposed ML-DRSNet and the improved ML-DCNN algorithms substantially enhance both the accuracy and real-time performance in decomposing MUSTs, establishing a technical foundation for neuro-information-driven motor intention recognition and disease assessment.CONCLUSIONSThe proposed ML-DRSNet and the improved ML-DCNN algorithms substantially enhance both the accuracy and real-time performance in decomposing MUSTs, establishing a technical foundation for neuro-information-driven motor intention recognition and disease assessment.
The swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time control in neural interfaces. However, the existing sEMG decomposition methods, including blind source separation (BSS) and deep learning, have not yet achieved satisfactory performance, due to high latency or low accuracy. This study introduces a novel real-time high-density sEMG (HD-sEMG) decomposition algorithm named ML-DRSNet, which combines multi-label learning with a deep residual shrinkage network (DRSNet) to improve accuracy and reduce latency. ML-DRSNet was evaluated on a public sEMG dataset and the corresponding MUSTs extracted via the convolutional BSS algorithm. An improved multi-label deep convolutional neural network (ML-DCNN) was also evaluated and compared against a conventional multi-task DCNN (MT-DCNN). These networks were trained and tested on various window sizes and step sizes. With the shortest window size (20 data points) and step size (10 data points), ML-DRSNet significantly outperformed both ML-DCNN (0.86 ± 0.18 vs. 0.71 ± 0.24, P < 0.001) and MT-DCNN (0.86 ± 0.18 vs. 0.66 ± 0.16, P < 0.001) in decomposition precision. Moreover, ML-DRSNet demonstrated a notably lower latency (15.15 ms) compared to ML-DCNN (69.36 ms) and MT-DCNN (76.96 ms), both of which demonstrated reduced latency relative to BSS-based decomposition methods. The proposed ML-DRSNet and the improved ML-DCNN algorithms substantially enhance both the accuracy and real-time performance in decomposing MUSTs, establishing a technical foundation for neuro-information-driven motor intention recognition and disease assessment.
ArticleNumber 106
Audience Academic
Author Ma, Jinting
Wang, Lifen
Dan, Guo
Zhang, Naiwen
Wei, Jing
Li, Jianjun
Jiang, Naifu
Tan, Lihai
Wu, Renxiang
Li, Qiuyuan
Li, Guanglin
Author_xml – sequence: 1
  givenname: Jinting
  surname: Ma
  fullname: Ma, Jinting
  organization: School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University
– sequence: 2
  givenname: Lifen
  surname: Wang
  fullname: Wang, Lifen
  organization: School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University
– sequence: 3
  givenname: Renxiang
  surname: Wu
  fullname: Wu, Renxiang
  organization: School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University
– sequence: 4
  givenname: Naiwen
  surname: Zhang
  fullname: Zhang, Naiwen
  organization: School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University
– sequence: 5
  givenname: Jing
  surname: Wei
  fullname: Wei, Jing
  organization: School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University
– sequence: 6
  givenname: Jianjun
  surname: Li
  fullname: Li, Jianjun
  organization: Rehabilitation Clinic, Shenzhen University General Hospital
– sequence: 7
  givenname: Qiuyuan
  surname: Li
  fullname: Li, Qiuyuan
  organization: Rehabilitation Clinic, Shenzhen University General Hospital
– sequence: 8
  givenname: Lihai
  surname: Tan
  fullname: Tan, Lihai
  organization: Guangdong-Hongkong-Macau Institute of CNS Regeneration and Jinan University (Shenzhen), Center for Language and Brain, Shenzhen Institute of Neuroscience
– sequence: 9
  givenname: Guanglin
  surname: Li
  fullname: Li, Guanglin
  organization: Guangdong-Hong Kong-Macao Joint Laboratory of Human-Machine Intelligence-Synergy Systems, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences
– sequence: 10
  givenname: Naifu
  surname: Jiang
  fullname: Jiang, Naifu
  email: nf.jiang@siat.ac.cn
  organization: Guangdong-Hong Kong-Macao Joint Laboratory of Human-Machine Intelligence-Synergy Systems, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences
– sequence: 11
  givenname: Guo
  surname: Dan
  fullname: Dan, Guo
  email: danguo@szu.edu.cn
  organization: School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University
BackLink https://www.ncbi.nlm.nih.gov/pubmed/40340912$$D View this record in MEDLINE/PubMed
BookMark eNp9kktv1DAURiNURB_wB1igSGzYpPiVjL1Co4qWSpXYwNpy7OuMp4492EnR_Hs8Tak6EkJZ5Orm3ONHvvPqJMQAVfUeo0uMefc5YyI4axBpG4Q7Khr6qjrDK0YbhBA9eVGfVuc5b0vBUMveVKfsUAlMzqq0rsfZT67xqgdfG4BdnSA7Mytf501y4V4NUAeYfsd0X9uY6o0bNo2BkN20r_OcrNJQgwc9pTju45DUbrMvJh3HXSyQi6F2oViVbyY3wtvqtVU-w7un90X18_rrj6tvzd33m9ur9V2jW0qnphU9YN1bSzhhYtVSDUq1oFcKKa6F0MyiriO97S1WlPZGKBAWDIeeGUE4vahuF6-Jait3yY0q7WVUTj42YhqkSpPTHqQhhHWC27YszVbICs4t7xlQ0iuhrCmuL4trN_cjGA1hSsofSY-_BLeRQ3yQmKCuXDYuhk9PhhR_zZAnObqswXsVIM5ZUoIIZnxFUUE_Luigyt5csLEo9QGXa04FYqxcSKEu_0GVx8DodEmKdaV_NPDh5RmeN_83DAUgC6BTzDmBfUYwkofEySVxsiROPiZO0jJEl6Fc4DBAkts4p1B-7P-m_gAXCNrd
Cites_doi 10.1088/1741-2560/11/1/016008
10.1088/0967-3334/35/7/R143
10.1152/jn.00452.2015
10.1109/JBHI.2022.3210019
10.1088/1741-2560/13/2/026027
10.1109/JBHI.2020.2987528
10.1088/1741-2552/ab2c55
10.1109/JBHI.2020.3041861
10.1088/1741-2552/ac5f1b
10.1088/1361-6579/aa60b9
10.1109/72.761722
10.1109/TAFFC.2022.3221554
10.1109/MSP.2014.2359242
10.1038/s41551-016-0025
10.1016/S0893-6080(00)00026-5
10.1109/TII.2019.2943898
10.1152/japplphysiol.00139.2019
10.1016/j.neuron.2008.10.037
10.1146/annurev.neuro.051508.135241
10.1088/1741-2552/abeead
10.1016/j.medengphy.2012.10.009
10.1152/japplphysiol.00635.2020
10.1523/ENEURO.0064-23.2023
10.1152/japplphysiol.00135.2018
10.1523/JNEUROSCI.0240-15.2015
10.1109/TBME.2020.2989311
10.1016/j.jelekin.2023.102796
10.3389/fnins.2024.1306054
10.1109/TBME.2020.3006508
10.1016/j.jelekin.2021.102548
10.1109/NER.2017.8008299
10.1109/TBME.2018.2817688
10.1109/TBME.2023.3294016
10.1152/japplphysiol.00453.2024
10.1109/TEM.2021.3104751
10.1016/j.jelekin.2014.08.014
10.1109/TPAMI.2022.3148324
10.1016/j.jelekin.2020.102426
10.1113/JP277396
ContentType Journal Article
Copyright The Author(s) 2025
2025. The Author(s).
COPYRIGHT 2025 BioMed Central Ltd.
The Author(s) 2025 2025
Copyright_xml – notice: The Author(s) 2025
– notice: 2025. The Author(s).
– notice: COPYRIGHT 2025 BioMed Central Ltd.
– notice: The Author(s) 2025 2025
DBID C6C
AAYXX
CITATION
CGR
CUY
CVF
ECM
EIF
NPM
7X8
5PM
DOA
DOI 10.1186/s12984-025-01639-3
DatabaseName Springer Nature OA Free Journals
CrossRef
Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
PubMed Central (Full Participant titles)
DOAJ: Directory of Open Access Journal (DOAJ)
DatabaseTitle CrossRef
MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList



MEDLINE - Academic
MEDLINE
Database_xml – sequence: 1
  dbid: DOA
  name: DOAJ: Directory of Open Access Journals
  url: https://www.doaj.org/
  sourceTypes: Open Website
– sequence: 2
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 3
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Medicine
Engineering
Occupational Therapy & Rehabilitation
Physical Therapy
EISSN 1743-0003
EndPage 19
ExternalDocumentID oai_doaj_org_article_d224698f5c53470f988f8b4e32ba9afd
PMC12060341
A839044249
40340912
10_1186_s12984_025_01639_3
Genre Journal Article
GeographicLocations Massachusetts
China
GeographicLocations_xml – name: Massachusetts
– name: China
GrantInformation_xml – fundername: Basic and Applied Basic Research Foundation of Guangdong Province
  grantid: 2024A1515012012
  funderid: http://dx.doi.org/10.13039/501100021171
– fundername: Shenzhen Medical Research Fund
  grantid: A2302006
– fundername: Medicine Plus Program of Shenzhen University
  grantid: 2024YG002
– fundername: Shenzhen Science and Technology Program
  grantid: KQTD20210811090217009
– fundername: National Natural Science Foundation of China
  grantid: 62001463; 61973220
  funderid: http://dx.doi.org/10.13039/501100001809
– fundername: National Natural Science Foundation of China
  grantid: 61973220
– fundername: Basic and Applied Basic Research Foundation of Guangdong Province
  grantid: 2024A1515012012
– fundername: National Natural Science Foundation of China
  grantid: 62001463
GroupedDBID ---
0R~
29L
2QV
2WC
53G
5GY
5VS
7RV
7X7
88E
8FE
8FG
8FH
8FI
8FJ
AAFWJ
AAJSJ
AASML
AAWTL
ABDBF
ABJCF
ABUWG
ACGFO
ACGFS
ACIWK
ACPRK
ACUHS
ADBBV
ADRAZ
ADUKV
AENEX
AFKRA
AFPKN
AFRAH
AHBYD
AHMBA
AHYZX
ALMA_UNASSIGNED_HOLDINGS
AMKLP
AMTXH
AOIJS
AQUVI
BAPOH
BAWUL
BBNVY
BCNDV
BENPR
BFQNJ
BGLVJ
BHPHI
BMC
BPHCQ
BVXVI
C6C
CCPQU
CS3
DIK
DU5
E3Z
EBD
EBLON
EBS
ESX
F5P
FYUFA
GROUPED_DOAJ
GX1
HCIFZ
HMCUK
HYE
I-F
IAO
IHR
INH
INR
IPY
ITC
KQ8
L6V
LK8
M0T
M1P
M7P
M7S
ML0
M~E
NAPCQ
O5R
O5S
OK1
OVT
P2P
PGMZT
PHGZM
PHGZT
PIMPY
PJZUB
PPXIY
PQGLB
PQQKQ
PROAC
PSQYO
PTHSS
PUEGO
RBZ
RNS
ROL
RPM
RSV
SBL
SOJ
TR2
TUS
UKHRP
WOQ
WOW
XSB
~8M
AAYXX
AFFHD
CITATION
ALIPV
CGR
CUY
CVF
ECM
EIF
NPM
7X8
5PM
ID FETCH-LOGICAL-c533t-59be1cbff28249753ceaa5ec7a0a8c99c4f0662bfbf1a33bd9ae9fed8eb4d9283
IEDL.DBID RSV
ISICitedReferencesCount 1
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001485265700002&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1743-0003
IngestDate Fri Oct 03 12:39:48 EDT 2025
Tue Nov 04 02:04:20 EST 2025
Fri Sep 05 17:15:59 EDT 2025
Sat Nov 29 13:49:07 EST 2025
Sat Nov 29 10:29:29 EST 2025
Mon May 12 02:38:39 EDT 2025
Sat Nov 29 07:56:37 EST 2025
Sat Sep 06 07:18:50 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 1
Keywords Motor unit spike trains
Multi-label learning
Real-time decomposition
High-density surface electromyography
Deep residual shrinkage network
Language English
License 2025. The Author(s).
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c533t-59be1cbff28249753ceaa5ec7a0a8c99c4f0662bfbf1a33bd9ae9fed8eb4d9283
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
OpenAccessLink https://link.springer.com/10.1186/s12984-025-01639-3
PMID 40340912
PQID 3202148730
PQPubID 23479
PageCount 19
ParticipantIDs doaj_primary_oai_doaj_org_article_d224698f5c53470f988f8b4e32ba9afd
pubmedcentral_primary_oai_pubmedcentral_nih_gov_12060341
proquest_miscellaneous_3202148730
gale_infotracmisc_A839044249
gale_infotracacademiconefile_A839044249
pubmed_primary_40340912
crossref_primary_10_1186_s12984_025_01639_3
springer_journals_10_1186_s12984_025_01639_3
PublicationCentury 2000
PublicationDate 2025-05-08
PublicationDateYYYYMMDD 2025-05-08
PublicationDate_xml – month: 05
  year: 2025
  text: 2025-05-08
  day: 08
PublicationDecade 2020
PublicationPlace London
PublicationPlace_xml – name: London
– name: England
PublicationTitle Journal of neuroengineering and rehabilitation
PublicationTitleAbbrev J NeuroEngineering Rehabil
PublicationTitleAlternate J Neuroeng Rehabil
PublicationYear 2025
Publisher BioMed Central
BioMed Central Ltd
BMC
Publisher_xml – name: BioMed Central
– name: BioMed Central Ltd
– name: BMC
References S Balasubramanian (1639_CR5) 2018; 65
D Farina (1639_CR9) 2017; 1
A Del Vecchio (1639_CR38) 2020; 53
Z Yang (1639_CR6) 2022; 19
H Zhao (1639_CR12) 2024; 71
A Hyvärinen (1639_CR17) 2000; 13
N Francesco (1639_CR15) 2016; 13
JC Kline (1639_CR25) 2016; 115
D Farina (1639_CR4) 2023; 71
1639_CR13
AK Clarke (1639_CR20) 2021; 68
1639_CR35
F Urh (1639_CR21) 2020; 2020
1639_CR36
A Holobar (1639_CR31) 2014; 11
1639_CR11
F Hug (1639_CR27) 2021; 130
1639_CR10
A Del Vecchio (1639_CR40) 2019; 597
Y Xu (1639_CR7) 2023; 27
X Du (1639_CR23) 2023; 14
1639_CR16
A Holobar (1639_CR28) 2014; 35
C Lin (1639_CR41) 2024; 18
CM Laine (1639_CR32) 2015; 35
X Zhang (1639_CR33) 2013; 35
A Del Vecchio (1639_CR30) 2019; 127
A Del Vecchio (1639_CR37) 2018; 125
1639_CR1
M Genzel (1639_CR19) 2023; 45
MS Tenan (1639_CR39) 2014; 24
D Farina (1639_CR3) 2015; 32
M Gazzoni (1639_CR43) 2017; 38
1639_CR42
C Chen (1639_CR14) 2020; 67
Y Zheng (1639_CR18) 2019; 16
1639_CR22
MG Ravi (1639_CR24) 2023; 70
M Zhao (1639_CR26) 2020; 16
Y Yu (1639_CR8) 2021; 25
S Ma (1639_CR34) 2021; 25
NG Hatsopoulos (1639_CR2) 2009; 32
F Hug (1639_CR29) 2021; 58
References_xml – volume: 11
  start-page: 016008
  year: 2014
  ident: 1639_CR31
  publication-title: J Neural Eng
  doi: 10.1088/1741-2560/11/1/016008
– volume: 35
  start-page: R143
  year: 2014
  ident: 1639_CR28
  publication-title: Physiol Meas
  doi: 10.1088/0967-3334/35/7/R143
– volume: 115
  start-page: 178
  year: 2016
  ident: 1639_CR25
  publication-title: J Neurophysiol
  doi: 10.1152/jn.00452.2015
– volume: 27
  start-page: 286
  year: 2023
  ident: 1639_CR7
  publication-title: IEEE J Biomed Health Inf
  doi: 10.1109/JBHI.2022.3210019
– volume: 13
  start-page: 026027
  year: 2016
  ident: 1639_CR15
  publication-title: J Neural Eng
  doi: 10.1088/1741-2560/13/2/026027
– volume: 25
  start-page: 47
  year: 2021
  ident: 1639_CR34
  publication-title: IEEE J Biomedical Health Inf
  doi: 10.1109/JBHI.2020.2987528
– volume: 16
  start-page: 066006
  year: 2019
  ident: 1639_CR18
  publication-title: J Neural Eng
  doi: 10.1088/1741-2552/ab2c55
– volume: 25
  start-page: 2557
  year: 2021
  ident: 1639_CR8
  publication-title: IEEE J Biomed Health Inf
  doi: 10.1109/JBHI.2020.3041861
– volume: 19
  start-page: 026030
  year: 2022
  ident: 1639_CR6
  publication-title: J Neural Eng
  doi: 10.1088/1741-2552/ac5f1b
– volume: 38
  start-page: R27
  year: 2017
  ident: 1639_CR43
  publication-title: Physiol Meas
  doi: 10.1088/1361-6579/aa60b9
– ident: 1639_CR16
  doi: 10.1109/72.761722
– volume: 14
  start-page: 2925
  year: 2023
  ident: 1639_CR23
  publication-title: IEEE Trans Affect Comput
  doi: 10.1109/TAFFC.2022.3221554
– ident: 1639_CR35
– volume: 32
  start-page: 115
  year: 2015
  ident: 1639_CR3
  publication-title: IEEE Signal Process Mag
  doi: 10.1109/MSP.2014.2359242
– volume: 1
  start-page: 0025
  year: 2017
  ident: 1639_CR9
  publication-title: Nat Biomed Eng
  doi: 10.1038/s41551-016-0025
– volume: 13
  start-page: 411
  year: 2000
  ident: 1639_CR17
  publication-title: Neural Netw
  doi: 10.1016/S0893-6080(00)00026-5
– volume: 16
  start-page: 4681
  year: 2020
  ident: 1639_CR26
  publication-title: IEEE Trans Ind Inf
  doi: 10.1109/TII.2019.2943898
– volume: 127
  start-page: 205
  year: 2019
  ident: 1639_CR30
  publication-title: J Appl Physiol
  doi: 10.1152/japplphysiol.00139.2019
– volume: 2020
  start-page: 736
  year: 2020
  ident: 1639_CR21
  publication-title: IEEE EMBC
– ident: 1639_CR1
  doi: 10.1016/j.neuron.2008.10.037
– volume: 32
  start-page: 249
  year: 2009
  ident: 1639_CR2
  publication-title: Annu Rev Neurosci
  doi: 10.1146/annurev.neuro.051508.135241
– ident: 1639_CR22
  doi: 10.1088/1741-2552/abeead
– volume: 35
  start-page: 537
  year: 2013
  ident: 1639_CR33
  publication-title: Med Eng Phys
  doi: 10.1016/j.medengphy.2012.10.009
– volume: 130
  start-page: 342
  year: 2021
  ident: 1639_CR27
  publication-title: J Appl Physiol
  doi: 10.1152/japplphysiol.00635.2020
– ident: 1639_CR10
  doi: 10.1523/ENEURO.0064-23.2023
– volume: 125
  start-page: 1404
  year: 2018
  ident: 1639_CR37
  publication-title: J Appl Physiol
  doi: 10.1152/japplphysiol.00135.2018
– volume: 35
  start-page: 12207
  year: 2015
  ident: 1639_CR32
  publication-title: J Neurosci
  doi: 10.1523/JNEUROSCI.0240-15.2015
– ident: 1639_CR36
– volume: 67
  start-page: 3501
  year: 2020
  ident: 1639_CR14
  publication-title: IEEE Trans Biomed Eng
  doi: 10.1109/TBME.2020.2989311
– volume: 71
  start-page: 102796
  year: 2023
  ident: 1639_CR4
  publication-title: J Electromyogr Kinesiol
  doi: 10.1016/j.jelekin.2023.102796
– volume: 18
  start-page: 1306054
  year: 2024
  ident: 1639_CR41
  publication-title: Front Neurosci
  doi: 10.3389/fnins.2024.1306054
– volume: 68
  start-page: 526
  year: 2021
  ident: 1639_CR20
  publication-title: IEEE Trans Biomed Eng
  doi: 10.1109/TBME.2020.3006508
– volume: 58
  start-page: 102548
  year: 2021
  ident: 1639_CR29
  publication-title: J Electromyogr Kinesiol
  doi: 10.1016/j.jelekin.2021.102548
– ident: 1639_CR42
  doi: 10.1109/NER.2017.8008299
– volume: 65
  start-page: 2790
  year: 2018
  ident: 1639_CR5
  publication-title: IEEE Trans Biomed Eng
  doi: 10.1109/TBME.2018.2817688
– volume: 71
  start-page: 160
  year: 2024
  ident: 1639_CR12
  publication-title: IEEE Trans Biomed Eng
  doi: 10.1109/TBME.2023.3294016
– ident: 1639_CR11
  doi: 10.1152/japplphysiol.00453.2024
– ident: 1639_CR13
– volume: 70
  start-page: 2787
  year: 2023
  ident: 1639_CR24
  publication-title: IEEE Trans Eng Manage
  doi: 10.1109/TEM.2021.3104751
– volume: 24
  start-page: 917
  year: 2014
  ident: 1639_CR39
  publication-title: J Electromyogr Kinesiol
  doi: 10.1016/j.jelekin.2014.08.014
– volume: 45
  start-page: 1119
  year: 2023
  ident: 1639_CR19
  publication-title: IEEE Trans Pattern Anal Mach Intell
  doi: 10.1109/TPAMI.2022.3148324
– volume: 53
  start-page: 102426
  year: 2020
  ident: 1639_CR38
  publication-title: J Electromyogr Kinesiol
  doi: 10.1016/j.jelekin.2020.102426
– volume: 597
  start-page: 2445
  year: 2019
  ident: 1639_CR40
  publication-title: J Physiol
  doi: 10.1113/JP277396
SSID ssj0034054
Score 2.40945
Snippet Background The swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time...
The swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time control in...
Background The swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time...
Abstract Background The swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling...
SourceID doaj
pubmedcentral
proquest
gale
pubmed
crossref
springer
SourceType Open Website
Open Access Repository
Aggregation Database
Index Database
Publisher
StartPage 106
SubjectTerms Algorithms
Biomedical and Life Sciences
Biomedical Engineering and Bioengineering
Biomedicine
Deep Learning
Deep residual shrinkage network
Electromyography
Electromyography - methods
High-definition television
High-density surface electromyography
Humans
Machine learning
Methods
Motor unit spike trains
Multi-label learning
Muscle, Skeletal - physiology
Neural Networks, Computer
Neurology
Neurophysiology
Neurosciences
Real-time decomposition
Rehabilitation Medicine
Signal processing
Signal Processing, Computer-Assisted
Specific gravity
SummonAdditionalLinks – databaseName: DOAJ: Directory of Open Access Journal (DOAJ)
  dbid: DOA
  link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1LbxMxEB6hCiE4FAivhYKMxOMAVjdZZ9c-BkTFhapCRerN8lONKNsqm1Tqv2fG662yRYILUk6xs_F6vnnJ428A3qDXCtHTmWETSi6aYLitbc1nFr1ZWVslRUzNJprDQ3lyoo62Wn1RTVhPD9xv3L4nxjMl49zNK9GUUUkZpRWhmlmjTPRkfctGDclUb4MrDEPEcEVG1vsdejUpOLVuxRCnUrwauaHE1v-nTd5ySjcLJm-cmiZndPAAdnMUyRb96h_CrdBO4N4Wt-AE7nzLp-YTeLtNJcyOex4B9o59H7F0T-D-URbaMOcRrBYslRxyBEs4Yz6EC4YJerrBxbpT_KufaJBY2xeTM4yAGREgc0918esr1m1W0bjAcredX1eZIRufRMXsuWKMLVt8qjnj1On-Mfw4-HL8-SvPbRo4iqRa87myYepsjJi9Cbqn64Ix8-AaUxrplHIiEs28jTZOTVVZT3zgMXgZrPAKw5snsNOet-EZMIxFvCVOJIxrhAhG-kb66GSIGGjVZVXAh0Fq-qJn49Api5G17mWsUcY6yVjj7E8k2OuZxKSdvkB86Ywv_S98FfCeYKFJ31H2zuRrC7hgYs7SC4wwSyHwzQvYG81EPXWj4dcDsDQNUXFbG843naYW9piVoq0t4GkPtOs1C8Kyms4KkCMIjl5qPNIuTxNN-BR1DX89LeDjgFadDVT3l117_j927QXcnSVtw4_cg531ahNewm13uV52q1dJV38DenZE7A
  priority: 102
  providerName: Directory of Open Access Journals
Title A multi-label deep residual shrinkage network for high-density surface electromyography decomposition in real-time
URI https://link.springer.com/article/10.1186/s12984-025-01639-3
https://www.ncbi.nlm.nih.gov/pubmed/40340912
https://www.proquest.com/docview/3202148730
https://pubmed.ncbi.nlm.nih.gov/PMC12060341
https://doaj.org/article/d224698f5c53470f988f8b4e32ba9afd
Volume 22
WOSCitedRecordID wos001485265700002&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVADU
  databaseName: BioMed Central Open Access Free
  customDbUrl:
  eissn: 1743-0003
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0034054
  issn: 1743-0003
  databaseCode: RBZ
  dateStart: 20040101
  isFulltext: true
  titleUrlDefault: https://www.biomedcentral.com/search/
  providerName: BioMedCentral
– providerCode: PRVAON
  databaseName: DOAJ: Directory of Open Access Journals
  customDbUrl:
  eissn: 1743-0003
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0034054
  issn: 1743-0003
  databaseCode: DOA
  dateStart: 20040101
  isFulltext: true
  titleUrlDefault: https://www.doaj.org/
  providerName: Directory of Open Access Journals
– providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  customDbUrl:
  eissn: 1743-0003
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0034054
  issn: 1743-0003
  databaseCode: M~E
  dateStart: 20040101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
– providerCode: PRVPQU
  databaseName: Biological Science Database (ProQuest)
  customDbUrl:
  eissn: 1743-0003
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0034054
  issn: 1743-0003
  databaseCode: M7P
  dateStart: 20090101
  isFulltext: true
  titleUrlDefault: http://search.proquest.com/biologicalscijournals
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Engineering Database
  customDbUrl:
  eissn: 1743-0003
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0034054
  issn: 1743-0003
  databaseCode: M7S
  dateStart: 20090101
  isFulltext: true
  titleUrlDefault: http://search.proquest.com
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Health & Medical Collection (ProQuest)
  customDbUrl:
  eissn: 1743-0003
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0034054
  issn: 1743-0003
  databaseCode: 7X7
  dateStart: 20090101
  isFulltext: true
  titleUrlDefault: https://search.proquest.com/healthcomplete
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Healthcare Administration Database (ProQuest)
  customDbUrl:
  eissn: 1743-0003
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0034054
  issn: 1743-0003
  databaseCode: M0T
  dateStart: 20090101
  isFulltext: true
  titleUrlDefault: https://search.proquest.com/healthmanagement
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Nursing & Allied Health Database
  customDbUrl:
  eissn: 1743-0003
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0034054
  issn: 1743-0003
  databaseCode: 7RV
  dateStart: 20090101
  isFulltext: true
  titleUrlDefault: https://search.proquest.com/nahs
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: ProQuest Central
  customDbUrl:
  eissn: 1743-0003
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0034054
  issn: 1743-0003
  databaseCode: BENPR
  dateStart: 20090101
  isFulltext: true
  titleUrlDefault: https://www.proquest.com/central
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Publicly Available Content Database
  customDbUrl:
  eissn: 1743-0003
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0034054
  issn: 1743-0003
  databaseCode: PIMPY
  dateStart: 20090101
  isFulltext: true
  titleUrlDefault: http://search.proquest.com/publiccontent
  providerName: ProQuest
– providerCode: PRVAVX
  databaseName: SpringerLINK Contemporary 1997-Present
  customDbUrl:
  eissn: 1743-0003
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0034054
  issn: 1743-0003
  databaseCode: RSV
  dateStart: 20041201
  isFulltext: true
  titleUrlDefault: https://link.springer.com/search?facet-content-type=%22Journal%22
  providerName: Springer Nature
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3db9MwED-xDSF4YFC-AqMyEh8PEJE0bmI_dmgTPLSKujKVJ8uObVaxpVPTIu2_5-wkUzMQEkjVPdSXD_c-XZ9_B_Aao5ax2u0ZZiYKaWZkqFKVhgOF0SxKFWfU-mYT2WTC5nOeN4fCqrbavd2S9J7amzVLP1YYmRgNXftVTFMSHiY7sIfhjrmGDdOT09b_JpiC0PZ4zB-v64Qgj9T_uz_eCkg3iyVv7Jj6QHS8_39TeAD3m8STjGpNeQi3TNmDe1twhD24M2422nvwZht9mMxq6AHylkw7wN492M8bObc8j2A1Ir5KMUT9MudEG3NJcE3vD32R6gwf9QN9GCnr-nOCSTNxmMmhdqX06ytSbVZWFoY0DXourhpQbbyTq39viszIosS7yvNwvbgwj-Hr8dHs0-ew6ewQFphersMhVyYulLW44KPuaG9hpByaIpORZAXnBbUOmV5ZZWOZJEo7CHFrNDOKao4Z0RPYLZeleQYE0xetHIwSpkKUGsl0xrQtmLGYm6VREsD7VtjisgbwEH7hw1JRy0OgPISXh0DuQ6cP15wOfNt_sVx9F40tC-1A-DizQ5wMzSLLGbNMUZMMlOTS6gDeOW0SzkWgyhSyOemAL-zAtsQIk9KIUpx5AAcdTjTtojP8qtVH4YZcPVxplptKuK73uJBF9xzA01o_r9-ZOhPg8SAA1tHczqS6I-XizCOLx2ieeHUcwIdWgUXj06q__GrP_439BdwdeBvADzuA3fVqY17C7eLnelGt-rCTTU8dnWeesj7sHR5N8mnf_1-CdBzNHM1yT09wPP8yzr_1vRv4BfWpUtU
linkProvider Springer Nature
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1bb9MwFD6Cgbg8MCi3wAAjcXkAi7RxE-exIKYhtmoaBe3N8pVVbOnUpEj79xw7ztQMhARSnuqTi3tun-VzPgO8wKxlnfF7hoVNKSuspCpXOR0pzGZprkrOXDhsophO-eFhuR-bwuqu2r3bkgyROrg1z9_VmJk4o_74VYQpWUmzy3CFYcbyjPkHX7518TdDCMK69pg_3tdLQYGp__d4vJaQLhZLXtgxDYloe_P_pnAbbkXgSSatpdyBS7YawM01OsIBXNuLG-0DeLnOPkxmLfUAeUUOesTeA9jcj3ruZO7CckJClSJF-7LHxFh7SnBNH5q-SH2Er_qBMYxUbf05QdBMPGcyNb6Uvjkj9WrppLYkHtBzchZJtfFJvv49FpmReYVPlce0mZ_Ye_B1--Psww6NJztQjfCyoeNS2aFWzuGCj_nWXm2lHFtdyFRyXZaaOc9Mr5xyQ5llyngKcWcNt4qZEhHRfdioFpV9CAThi1GeRgmhEGNWclNw4zS3DrFZnmYJvOmULU5bAg8RFj48F60-BOpDBH0IlH7v7eFc0pNvhx8Wy-8i-rIwnoSv5G6Mk2FF6krOHVfMZiMlS-lMAq-9NQkfItBktIydDvjBnmxLTBCUpozhzBPY6kmia-ve8PPOHoUf8vVwlV2sauFPvceFLIbnBB609nn-zcy7QDkcJcB7ltubVH-kmh8FZvEhuifePUzgbWfAIsa0-i__2qN_E38G13dme7ti99P082O4MQr-gBffgo1mubJP4Kr-2czr5dPg2L8Az3lLyg
linkToPdf http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1bb9MwFD6CgSZ4YFBugQFG4vIA1tLGTezHcqlAQDXBQHuzfGUVW1o1KdL-PcdOMjUDISGkPNUnF_fcPsvnfAZ4glnLeRv2DAuXUlY4RXWuczrSmM3SXAvOfDxsopjN-OGh2N_o4o_V7t2WZNPTEFiaynpvaX3j4jzfqzBLcUbDUawIWTJBs4twiYVC-rBe__Kti8UZwhHWtcr88b5eOoqs_b_H5o3kdL5w8tzuaUxK053_n851uNYCUjJpLOgGXHDlAK5u0BQOYPtTuwE_gKebrMTkoKEkIM_I5x7h9wB29lv9dzI3YTUhsXqRot25Y2KdWxJc68dmMFId4at-YGwjZVOXThBMk8ClTG0osa9PSbVeeWUcaQ_uOTltybbxSaEuvi0-I_MSn6qOaT0_cbfg6_Ttwet3tD3xgRqEnTUdC-2GRnuPC0EWWn6NU2rsTKFSxY0QhvnAWK-99kOVZdoGanHvLHeaWYFI6TZslYvS3QWCsMbqQK-EEIkxp7gtuPWGO4-YLU-zBF50ipfLhthDxgURz2WjD4n6kFEfEqVfBds4kwyk3PGHxeq7bH1c2kDOJ7gf42RYkXrBueeauWyklVDeJvA8WJYMoQPNx6i2AwI_OJBwyQmC1ZQxnHkCuz1JdHnTG37c2aYMQ6FOrnSLdSWzUeDA4xi2E7jT2OrZN7PgDmI4SoD3rLg3qf5IOT-KjONDdFu8e5jAy86YZRvrqr_8a_f-TfwRbO-_mcqP72cf7sOVUXQHvPgubNWrtXsAl83Pel6tHkYf_wXOM1Su
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=A+multi-label+deep+residual+shrinkage+network+for+high-density+surface+electromyography+decomposition+in+real-time&rft.jtitle=Journal+of+neuroengineering+and+rehabilitation&rft.au=Ma%2C+Jinting&rft.au=Wang%2C+Lifen&rft.au=Wu%2C+Renxiang&rft.au=Zhang%2C+Naiwen&rft.date=2025-05-08&rft.pub=BioMed+Central&rft.eissn=1743-0003&rft.volume=22&rft.issue=1&rft_id=info:doi/10.1186%2Fs12984-025-01639-3&rft.externalDocID=10_1186_s12984_025_01639_3
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1743-0003&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1743-0003&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1743-0003&client=summon