Presentation 2020-11-04
Base-type identification model for next-generation DNA sequencer using CNN
Daisuke Hayashi, Toru Yokoyama, Kiyohiro Obara,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Next-generation DNA sequencers are expected to grow in the market as their applications in the medical field such as cancer diagnosis and personalized medicine. In the conventional model, since the base types of DNA are identified only using a fluorescence intensity at each DNA colony position, their accuracies are easily affected by image defocus and bad registration. Generally, it is difficult to model crosstalk among fluorescent signals and phasing correction between cycles. In this study, we proposed a base discrimination model based on CNN, which improved performance by iterating CNN learning 5 times. As a result, we confirmed the effectiveness of CNN iterative learning and obtained the prospect of its device-independence, which means that it will not be necessary to optimize a model for each device.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Next-generation DNA sequencer / CNN / Iterative learning
Paper # MICT2020-10,MI2020-36
Date of Issue 2020-10-28 (MICT, MI)

Conference Information
Committee MICT / MI
Conference Date 2020/11/4(1days)
Place (in Japanese) (See Japanese page)
Place (in English) Virtual (TBD)
Topics (in Japanese) (See Japanese page)
Topics (in English) Medical imaging technology, healthcare and medical information communication technology
Chair Eisuke Hanada(Saga Univ.) / Yoshiki Kawata(Tokushima Univ.)
Vice Chair Hirokazu Tanaka(Hiroshima City Univ.) / Daisuke Anzai(Nagoya Inst. of Tech.) / Takayuki Kitasaka(Aichi Inst. of Tech.) / Hidekata Hontani(Nagoya Inst. of Tech.)
Secretary Hirokazu Tanaka(Kobe Univ.) / Daisuke Anzai(Yokohama National Univ.) / Takayuki Kitasaka(Yamaguchi Univ.) / Hidekata Hontani(Univ. of Hyogo)
Assistant Keita Saku(Kyushu Univ.) / Kai Ishida(KISTEC) / Kento Takabayashi(Okayama Pref. Univ.) / Hotaka Takizawa(Tsukuba Univ.) / Yoshito Otake(NAIST)

Paper Information
Registration To Technical Committee on Healthcare and Medical Information Communication Technology / Technical Committee on Medical Imaging
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Base-type identification model for next-generation DNA sequencer using CNN
Sub Title (in English)
Keyword(1) Next-generation DNA sequencer
Keyword(2) CNN
Keyword(3) Iterative learning
1st Author's Name Daisuke Hayashi
1st Author's Affiliation Hitachi, Ltd., Research & Development Group, Center for Technology Innovation - Artificial Intelligence(Hitachi)
2nd Author's Name Toru Yokoyama
2nd Author's Affiliation Hitachi, Ltd., Research & Development Group, Center for Technology Innovation - Artificial Intelligence(Hitachi)
3rd Author's Name Kiyohiro Obara
3rd Author's Affiliation Hitachi, Ltd., Research & Development Group, Center for Technology Innovation - Artificial Intelligence(Hitachi)
Date 2020-11-04
Paper # MICT2020-10,MI2020-36
Volume (vol) vol.120
Number (no) MICT-219,MI-220
Page pp.pp.15-20(MICT), pp.15-20(MI),
#Pages 6
Date of Issue 2020-10-28 (MICT, MI)