Recognizing aspiration presence using model parameter classification from microwave doppler signals

Shuhei Inui, Kosuke Okusa, Kurato Maeno, Toshinari Kanakura

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

A study on the healthcare application is very important for the solitary death in aging society. Many previous works had been proposed a detection method of aspiration using the non-contact radar. But the works are only in subjects with sitting in a chair. We consider that user falls down in the state when he happen abnormal situation as daily life. In this study, we focus on the detection of "aspiration" or "apnea" for the lying position, because the final decision of the life or death is aspiration. As initial stage of the system, we propose the recognition method for the presence of aspiration with lying position under the low-disturbance environment from microwave Doppler signals by using support vector machine (SVM).

Original languageEnglish
Title of host publicationInternational MultiConference of Engineers and Computer Scientists, IMECS 2012
EditorsJon Burgstone, S. I. Ao, Craig Douglas, W. S. Grundfest
PublisherNewswood Limited
Pages509-512
Number of pages4
ISBN (Electronic)9789881925169
ISBN (Print)9789881925114
Publication statusPublished - Jan 1 2012
Externally publishedYes
Event2012 World Congress on Engineering and Computer Science, WCECS 2012 - San Francisco, United States
Duration: Oct 24 2012Oct 26 2012

Publication series

NameLecture Notes in Engineering and Computer Science
Volume1
ISSN (Print)2078-0958

Other

Other2012 World Congress on Engineering and Computer Science, WCECS 2012
CountryUnited States
CitySan Francisco
Period10/24/1210/26/12

All Science Journal Classification (ASJC) codes

  • Computer Science (miscellaneous)

Fingerprint Dive into the research topics of 'Recognizing aspiration presence using model parameter classification from microwave doppler signals'. Together they form a unique fingerprint.

Cite this