Presentation 2022-03-29
Classification of power spectra of EEG by non-negative matrix factorization
Kazuki Koyama, Mariko Ito, Masanori Sakaguchi, Takaaki Ohnishi,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) It has been common for experts to manually and subjectively classify sleep stages in the experiments for the sleep of mice. Recently, however, automized methods for classification using supervised learning have been proposed. We attempt to determine the sleep stage by an unsupervised learning, Non-negative Matrix Factorization. By applying Non-negative Matrix Factorization to a large matrix incorporating power spectra of EEG of all the mice for each time window, we obtained the basis and activation matrices, where the former matrix exhibits the bases of power spectra. We examined whether we can determine the sleep stage of mice by comparing the activation corresponding to each basis with the label of sleep stage given by experts. In addition, we, for each mouse, obtained the basis matrix by applying Non-negative Matrix Factorization to the matrix of power spectrum for each mouse. We classified mice based on the similarity of the bases of power spectra. The result implies the existence of universal bases of power spectra for mice.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Non-negative Matrix Factorization / EEG / power spectra / sleep stage classification / mouse
Paper # MSS2021-76,NLP2021-147
Date of Issue 2022-03-21 (MSS, NLP)

Conference Information
Committee MSS / NLP
Conference Date 2022/3/28(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Online
Topics (in Japanese) (See Japanese page)
Topics (in English) MSS, NLP, Work In Progress (MSS only), and etc.
Chair Atsuo Ozaki(Osaka Inst. of Tech.) / Takuji Kosaka(Chukyo Univ.)
Vice Chair Shingo Yamaguchi(Yamaguchi Univ.) / Akio Tsuneda(Kumamoto Univ.)
Secretary Shingo Yamaguchi(Hokkaido Univ.) / Akio Tsuneda(NEC)
Assistant Masato Shirai(Shimane Univ.) / Hideyuki Kato(Oita Univ.) / Yuichi Yokoi(Nagasaki Univ.)

Paper Information
Registration To Technical Committee on Mathematical Systems Science and its Applications / Technical Committee on Nonlinear Problems
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Classification of power spectra of EEG by non-negative matrix factorization
Sub Title (in English)
Keyword(1) Non-negative Matrix Factorization
Keyword(2) EEG
Keyword(3) power spectra
Keyword(4) sleep stage classification
Keyword(5) mouse
1st Author's Name Kazuki Koyama
1st Author's Affiliation Rikkyo University(Rikkyo Univ.)
2nd Author's Name Mariko Ito
2nd Author's Affiliation Rikkyo University(Rikkyo Univ.)
3rd Author's Name Masanori Sakaguchi
3rd Author's Affiliation University of Tsukuba(Univ. Tsukuba)
4th Author's Name Takaaki Ohnishi
4th Author's Affiliation Rikkyo University(Rikkyo Univ.)
Date 2022-03-29
Paper # MSS2021-76,NLP2021-147
Volume (vol) vol.121
Number (no) MSS-443,NLP-444
Page pp.pp.111-116(MSS), pp.111-116(NLP),
#Pages 6
Date of Issue 2022-03-21 (MSS, NLP)