Incremental Learning Method for Biological Signal Identification

Tadahiro Oyama, Stephen Karungaru, Satoru Tsuge, Yasue Mitsukura, Minoru Fukumi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

There is electromyogram(EMG) as one of biological signals generated along with motions of the human body. This EMG has information corresponding to condition of power, tenderness of motions and motion level. Therefore, it is thought that it is useful biological information for analyzing person's motions. Recently, researches on this EMG have been actively done. For instance, it is used as a control signal of electrical arms, because EMG can be gathered from the remaining muscle about upper-extremity amputee. In addition, the pointing device that uses EMG has been developed. In general, EMG is measured from a part with comparatively big muscular fibers such as arms and shoulders. There is a problem that placing and removing the electrode is inconvenient when EMG is measured by placing the electrode in the arm and the shoulder. Therefore, if we can recognize wrist motions using EMG which is measured from the wrist, the range of application will extend furthermore. Currently, we have constructed a wrist motion recognition system which recognizes the wrist motion of 7 types as an object by using the technique named Simple-FLDA as a feature extraction technique. However, motions with low recognition accuracy were observed, and it was found that the difference of the recognition accuracy is significant at each motion and subject. Because EMG is highly individual and its repeatability is low. Therefore, it is necessary to deal with these problems. In this paper, we try the construction of a system that can learn by giving incremental data to achieve an online tuning. The improvement of the algorithm of Simple-FLDA that incremental learning becomes possible was tried as a technique for the construction of online tuning system. As a recognition experimental result, we can confirm the rising of the recognition accuracy by incremental learning.

Original languageEnglish
Title of host publicationIFMBE Proceedings
Pages302-305
Number of pages4
Volume23
DOIs
Publication statusPublished - 2009
Externally publishedYes
Event13th International Conference on Biomedical Engineering, ICBME 2008 - , Singapore
Duration: 2008 Dec 32008 Dec 6

Other

Other13th International Conference on Biomedical Engineering, ICBME 2008
CountrySingapore
Period08/12/308/12/6

Fingerprint

Tuning
Electrodes
Muscle
Feature extraction
Fibers

Keywords

  • EMG
  • incremental learning
  • Incremental Simple-FLDA
  • Simple-FLDA

ASJC Scopus subject areas

  • Biomedical Engineering
  • Bioengineering

Cite this

Oyama, T., Karungaru, S., Tsuge, S., Mitsukura, Y., & Fukumi, M. (2009). Incremental Learning Method for Biological Signal Identification. In IFMBE Proceedings (Vol. 23, pp. 302-305) https://doi.org/10.1007/978-3-540-92841-6_74

Incremental Learning Method for Biological Signal Identification. / Oyama, Tadahiro; Karungaru, Stephen; Tsuge, Satoru; Mitsukura, Yasue; Fukumi, Minoru.

IFMBE Proceedings. Vol. 23 2009. p. 302-305.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Oyama, T, Karungaru, S, Tsuge, S, Mitsukura, Y & Fukumi, M 2009, Incremental Learning Method for Biological Signal Identification. in IFMBE Proceedings. vol. 23, pp. 302-305, 13th International Conference on Biomedical Engineering, ICBME 2008, Singapore, 08/12/3. https://doi.org/10.1007/978-3-540-92841-6_74
Oyama T, Karungaru S, Tsuge S, Mitsukura Y, Fukumi M. Incremental Learning Method for Biological Signal Identification. In IFMBE Proceedings. Vol. 23. 2009. p. 302-305 https://doi.org/10.1007/978-3-540-92841-6_74
Oyama, Tadahiro ; Karungaru, Stephen ; Tsuge, Satoru ; Mitsukura, Yasue ; Fukumi, Minoru. / Incremental Learning Method for Biological Signal Identification. IFMBE Proceedings. Vol. 23 2009. pp. 302-305
@inproceedings{4c4aec196eb14d9a9e8da626b1aefe1b,
title = "Incremental Learning Method for Biological Signal Identification",
abstract = "There is electromyogram(EMG) as one of biological signals generated along with motions of the human body. This EMG has information corresponding to condition of power, tenderness of motions and motion level. Therefore, it is thought that it is useful biological information for analyzing person's motions. Recently, researches on this EMG have been actively done. For instance, it is used as a control signal of electrical arms, because EMG can be gathered from the remaining muscle about upper-extremity amputee. In addition, the pointing device that uses EMG has been developed. In general, EMG is measured from a part with comparatively big muscular fibers such as arms and shoulders. There is a problem that placing and removing the electrode is inconvenient when EMG is measured by placing the electrode in the arm and the shoulder. Therefore, if we can recognize wrist motions using EMG which is measured from the wrist, the range of application will extend furthermore. Currently, we have constructed a wrist motion recognition system which recognizes the wrist motion of 7 types as an object by using the technique named Simple-FLDA as a feature extraction technique. However, motions with low recognition accuracy were observed, and it was found that the difference of the recognition accuracy is significant at each motion and subject. Because EMG is highly individual and its repeatability is low. Therefore, it is necessary to deal with these problems. In this paper, we try the construction of a system that can learn by giving incremental data to achieve an online tuning. The improvement of the algorithm of Simple-FLDA that incremental learning becomes possible was tried as a technique for the construction of online tuning system. As a recognition experimental result, we can confirm the rising of the recognition accuracy by incremental learning.",
keywords = "EMG, incremental learning, Incremental Simple-FLDA, Simple-FLDA",
author = "Tadahiro Oyama and Stephen Karungaru and Satoru Tsuge and Yasue Mitsukura and Minoru Fukumi",
year = "2009",
doi = "10.1007/978-3-540-92841-6_74",
language = "English",
isbn = "9783540928409",
volume = "23",
pages = "302--305",
booktitle = "IFMBE Proceedings",

}

TY - GEN

T1 - Incremental Learning Method for Biological Signal Identification

AU - Oyama, Tadahiro

AU - Karungaru, Stephen

AU - Tsuge, Satoru

AU - Mitsukura, Yasue

AU - Fukumi, Minoru

PY - 2009

Y1 - 2009

N2 - There is electromyogram(EMG) as one of biological signals generated along with motions of the human body. This EMG has information corresponding to condition of power, tenderness of motions and motion level. Therefore, it is thought that it is useful biological information for analyzing person's motions. Recently, researches on this EMG have been actively done. For instance, it is used as a control signal of electrical arms, because EMG can be gathered from the remaining muscle about upper-extremity amputee. In addition, the pointing device that uses EMG has been developed. In general, EMG is measured from a part with comparatively big muscular fibers such as arms and shoulders. There is a problem that placing and removing the electrode is inconvenient when EMG is measured by placing the electrode in the arm and the shoulder. Therefore, if we can recognize wrist motions using EMG which is measured from the wrist, the range of application will extend furthermore. Currently, we have constructed a wrist motion recognition system which recognizes the wrist motion of 7 types as an object by using the technique named Simple-FLDA as a feature extraction technique. However, motions with low recognition accuracy were observed, and it was found that the difference of the recognition accuracy is significant at each motion and subject. Because EMG is highly individual and its repeatability is low. Therefore, it is necessary to deal with these problems. In this paper, we try the construction of a system that can learn by giving incremental data to achieve an online tuning. The improvement of the algorithm of Simple-FLDA that incremental learning becomes possible was tried as a technique for the construction of online tuning system. As a recognition experimental result, we can confirm the rising of the recognition accuracy by incremental learning.

AB - There is electromyogram(EMG) as one of biological signals generated along with motions of the human body. This EMG has information corresponding to condition of power, tenderness of motions and motion level. Therefore, it is thought that it is useful biological information for analyzing person's motions. Recently, researches on this EMG have been actively done. For instance, it is used as a control signal of electrical arms, because EMG can be gathered from the remaining muscle about upper-extremity amputee. In addition, the pointing device that uses EMG has been developed. In general, EMG is measured from a part with comparatively big muscular fibers such as arms and shoulders. There is a problem that placing and removing the electrode is inconvenient when EMG is measured by placing the electrode in the arm and the shoulder. Therefore, if we can recognize wrist motions using EMG which is measured from the wrist, the range of application will extend furthermore. Currently, we have constructed a wrist motion recognition system which recognizes the wrist motion of 7 types as an object by using the technique named Simple-FLDA as a feature extraction technique. However, motions with low recognition accuracy were observed, and it was found that the difference of the recognition accuracy is significant at each motion and subject. Because EMG is highly individual and its repeatability is low. Therefore, it is necessary to deal with these problems. In this paper, we try the construction of a system that can learn by giving incremental data to achieve an online tuning. The improvement of the algorithm of Simple-FLDA that incremental learning becomes possible was tried as a technique for the construction of online tuning system. As a recognition experimental result, we can confirm the rising of the recognition accuracy by incremental learning.

KW - EMG

KW - incremental learning

KW - Incremental Simple-FLDA

KW - Simple-FLDA

UR - http://www.scopus.com/inward/record.url?scp=84891931552&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84891931552&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-92841-6_74

DO - 10.1007/978-3-540-92841-6_74

M3 - Conference contribution

AN - SCOPUS:84891931552

SN - 9783540928409

VL - 23

SP - 302

EP - 305

BT - IFMBE Proceedings

ER -