Presentation 2013-08-30
Yes/No Communication Method for ALS Patients Using Their Biosignals : A Comparative Study Between Discriminant Analysis Method and Support Vector Machine
Kuniaki OZAWA, Junko SENGOKU, Masayoshi NAITO, Isamu OGATA, Tsuneo KANAZAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) An Yes/No communication method is studied for ALS(Amyotrophic Lateral Sclerosis) patients who are unable to move their bodies. The method uses cerebral blood volume changes measured with near-infrared light. While Yes/No answers have been so far judged by a discriminant analysis (DA) that classifies blood volume waves into Yes/No groups, we propose in this report the use of SVM (support Vector Machine) to improve accuracy rates. We judged the waves of 17 subjects (11 healthy subjects and 6 ALS patients) for three tasks by DA and SVM. The result has shown that SVM has higher accuracy rate than DA with 5% significant level under the variables of oscillation number of blood volume change and that of heart rate. The following tasks for Yes/No have shown the significant difference; "12- second- activation and 12- second -rest" for Yes task and "24 -second -rest" and "3 times repetition of 4- second- activation and 4- second- rest" for No tasks.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) ALS / communication / Yes/No answers / near- infrared light / cerebral blood volume / discriminant analysis / support vector machine
Paper # WIT2013-35
Date of Issue

Conference Information
Committee WIT
Conference Date 2013/8/22(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Well-being Information Technology(WIT)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Yes/No Communication Method for ALS Patients Using Their Biosignals : A Comparative Study Between Discriminant Analysis Method and Support Vector Machine
Sub Title (in English)
Keyword(1) ALS
Keyword(2) communication
Keyword(3) Yes/No answers
Keyword(4) near- infrared light
Keyword(5) cerebral blood volume
Keyword(6) discriminant analysis
Keyword(7) support vector machine
1st Author's Name Kuniaki OZAWA
1st Author's Affiliation Division of Mathematical Sciences, Tokyo Woman's Christian University()
2nd Author's Name Junko SENGOKU
2nd Author's Affiliation Division of Mathematical Sciences, Tokyo Woman's Christian University
3rd Author's Name Masayoshi NAITO
3rd Author's Affiliation Division of Mathematical Sciences, Tokyo Woman's Christian University
4th Author's Name Isamu OGATA
4th Author's Affiliation Excel of Mechatronix, Co.
5th Author's Name Tsuneo KANAZAWA
5th Author's Affiliation Excel of Mechatronix, Co.
Date 2013-08-30
Paper # WIT2013-35
Volume (vol) vol.113
Number (no) 195
Page pp.pp.-
#Pages 6
Date of Issue