Walk detection with a kinematic sensor: Frequency and wavelet comparison

  • Pierre Barralon*
  • , Nicolas Vuillerme
  • , Norbert Noury
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

73 Citations (Scopus)

Abstract

This study is included in the framework of Health Smart Homes which monitor some physiological or not physiological parameters of elderly people living independently at home. In this study we will focus on the walk detection. Walk activity is one parameter to evaluate the health of patient. For example, the total time of walk during a day allows assessing quickly if the subject is mobile rather than immobile. To reach this goal we used a kinematic sensor placed on the chest recording the movements of the subject. The data are analyzed by six algorithms to detect walk phases: two based on Fourier analysis and the others using a wavelet decomposition (DWT and CWT). All algorithms are described and the performances are evaluated on real data recorded with 20 elderly people. Results show that the method using the DWT decomposition is the most efficient (78.5% in sensitivity and 67.6% in specificity).

Original languageEnglish
Title of host publication28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'06
Pages1711-1714
Number of pages4
DOIs
Publication statusPublished - 2006
Externally publishedYes
Event28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'06 - New York, NY, United States
Duration: 30 Aug 20063 Sept 2006

Publication series

NameAnnual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings
ISSN (Print)0589-1019

Conference

Conference28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'06
Country/TerritoryUnited States
CityNew York, NY
Period30/08/063/09/06

Fingerprint

Dive into the research topics of 'Walk detection with a kinematic sensor: Frequency and wavelet comparison'. Together they form a unique fingerprint.

Cite this