Presentation | 2019-11-05 A study of machine learning algorithm for wearable biosignal sensor Daisuke Watanabe, Yuji Yano, Shintaro Izumi, Hiroshi Kawaguchi, Masahiko Yosimoto, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The algorithm was evaluated assuming that edge inference was performed on the data obtained from the wearable biological information sensor. For three applications for wearable healthcare, we evaluated the algorithm from the viewpoint of inference accuracy and energy efficiency by implementing a random forest (RF) and convolutional neural network (CNN) with FPGA. As a result, RF increases energy efficiency by one to three orders of magnitude, making it suitable for low power applications. On the other hand, inferior accuracy, CNN is 3% to 10% high, so it is suitable for applications that require high accuracy. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | IoT / wearable healthcare / machine learning / low power inference |
Paper # | MICT2019-25,MI2019-52 |
Date of Issue | 2019-10-29 (MICT, MI) |
Conference Information | |
Committee | MI / MICT |
---|---|
Conference Date | 2019/11/5(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Univ. of Tsukuba |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Medical imaging technology, healthcare and medical information communication technology |
Chair | Yoshiki Kawata(Tokushima Univ.) / Shinsuke Hara(Osaka City Univ.) |
Vice Chair | Takayuki Kitasaka(Aichi Inst. of Tech.) / Hidekata Hontani(Nagoya Inst. of Tech.) / Eisuke Hanada(Saga Univ.) / Chika Sugimoto(Yokohama National Univ.) |
Secretary | Takayuki Kitasaka(Yamaguchi Univ.) / Hidekata Hontani(Univ. of Hyogo) / Eisuke Hanada(Nagoya Inst. of Tech.) / Chika Sugimoto(Kobe Univ.) |
Assistant | Hotaka Takizawa(Tsukuba Univ.) / Yoshito Otake(NAIST) / Takumi Kobayashi(Yokohama National Univ.) / Keita Saku(Kyushu Univ.) / Kai Ishida(NICT) / Kento Takabayashi(Okayama Pref. Univ.) |
Paper Information | |
Registration To | Technical Committee on Medical Imaging / Technical Committee on Healthcare and Medical Information Communication Technology |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | A study of machine learning algorithm for wearable biosignal sensor |
Sub Title (in English) | |
Keyword(1) | IoT |
Keyword(2) | wearable healthcare |
Keyword(3) | machine learning |
Keyword(4) | low power inference |
1st Author's Name | Daisuke Watanabe |
1st Author's Affiliation | Kobe University(Kobe Univ.) |
2nd Author's Name | Yuji Yano |
2nd Author's Affiliation | Kobe University(Kobe Univ.) |
3rd Author's Name | Shintaro Izumi |
3rd Author's Affiliation | Kobe University(Kobe Univ.) |
4th Author's Name | Hiroshi Kawaguchi |
4th Author's Affiliation | Kobe University(Kobe Univ.) |
5th Author's Name | Masahiko Yosimoto |
5th Author's Affiliation | Kobe University(Kobe Univ.) |
Date | 2019-11-05 |
Paper # | MICT2019-25,MI2019-52 |
Volume (vol) | vol.119 |
Number (no) | MICT-263,MI-264 |
Page | pp.pp.7-8(MICT), pp.7-8(MI), |
#Pages | 2 |
Date of Issue | 2019-10-29 (MICT, MI) |