Skip to Main content Skip to Navigation
Theses

Détection de motifs temporels dans les environnements multi-perceptifs. Application à la classification automatique des Activités de la Vie Quotidienne d'une personne suivie à domicile par télémédecine

Anthony Fleury 1
1 AFIRM
TIMC - Techniques de l'Ingénierie Médicale et de la Complexité - Informatique, Mathématiques et Applications, Grenoble - UMR 5525
Abstract : In the near 2050, about one third of the French population will be over 65. The works of the team AFIRM of the TIMC-IMAG focus on the monitoring of elderly people at home, to detect, as earlier as possible, a loss of autonomy. This thesis work aims at objectivising criterions as ADL or AGGIR grids by automatically classifying the different Activities of Daily Living performed by the subject during the day. A Health Smart Home is used to do this.
Our Smart Home includes, in a real flat, Presence Infra-Red Sensors, (for localisation), door contacts (for the use of some conveniences), temperature and hygrometry sensors in the bathroom and microphones (sound and speech recognition with the GETALP team of the LIG). The subject is also equipped with a kinematic sensor that delivers on the postural transition (by pattern recognition with wavelets) and walk periods (frequency analysis).
This manuscript is compound of two major parts. The first one introduces the realisation of the kinematic sensor, its algorithms and its first evaluation; but also the set-up, algorithms and the validation of the other sensors inside the flat and finally the Support Vector Machines algorithms to classify the activities of Daily Living (hygiene, toilets, preparing and having a meal, resting, sleeping, communication and dressing/undressing). The second part deals with the experimental protocol to validate these algorithms and the results of these validations on young and healthy subjects. It introduces the results and a discussion about their validity.
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-00336400
Contributor : Anthony Fleury <>
Submitted on : Tuesday, November 4, 2008 - 9:24:20 PM
Last modification on : Thursday, November 19, 2020 - 3:56:08 PM
Long-term archiving on: : Saturday, November 26, 2016 - 3:03:26 AM

Identifiers

  • HAL Id : tel-00336400, version 2

Collections

Citation

Anthony Fleury. Détection de motifs temporels dans les environnements multi-perceptifs. Application à la classification automatique des Activités de la Vie Quotidienne d'une personne suivie à domicile par télémédecine. Sciences de l'ingénieur [physics]. Université Joseph-Fourier - Grenoble I, 2008. Français. ⟨tel-00336400v2⟩

Share

Metrics

Record views

557

Files downloads

5150