Deep Learning-Based Water-Intake Estimation Method Using Second Half of Swallowing Sound

被引：0

作者：

Yamada, Yutaro ^{[1
]}

Nishimura, Masafumi ^{[1
]}

Mineno, Hiroshi ^{[1
]}

Saito, Takato ^{[2
]}

Kawasaki, Satoshi ^{[2
]}

Ikeda, Daizo ^{[2
]}

Katagiri, Masaji ^{[2
]}

机构：

[1] Shizuoka Univ, Grad Sch Integrated Sci & Technol, Naka Ku, 3-5-1 Johoku, Hamamatsu, Shizuoka 4328011, Japan

[2] NTT Docomo Inc, Urban Sensing Res Grp, Res Labs, Serv Dev Div 2,Serv Innovat Dept, 3-5 Hikari No Oka, Yokosuka, Kanagawa 2398536, Japan

来源：

2017 IEEE 6TH GLOBAL CONFERENCE ON CONSUMER ELECTRONICS (GCCE) | 2017年

关键词：

deep learning; water-intake estimation; swallowing sound; healthcare;

D O I：

暂无

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

"Super-aged" societies are facing a staggering shortage of nurses and caregivers. Although water-intake is a necessity regarding healthcare management of elderly people, it is not currently automated. Thus, it is a burden on caregivers. We investigated how to estimate water intake by analyzing swallowing sounds. However, the estimation error for each subject was large because of the difficulty of discovering and extracting the common features correlated with appropriate water intake for subjects from complicated swallowing sounds. We thus propose a deep learning-based water-intake estimation method using the second half of a swallowing sound, which is correlated with water-intake.

引用

页数：2