Skip to Main content Skip to Navigation

Fusion de Données Multicapteurs pour un Système de Télésurveillance Médicale de Personnes à Domicile

Florence Duchêne 1
TIMC - Techniques de l'Ingénierie Médicale et de la Complexité - Informatique, Mathématiques et Applications, Grenoble - UMR 5525
Abstract : The development of medical remote care applications is crucial due to the general aging of the population and the restricted number of possible admissions to hospital, residential or nursing homes. This thesis deals with the concept of smart assistants used to analyze large heterogeneous data sets collected at home from sensors, in order to detect and prevent unusual situations that may give serious cause for concern. The aim is to learn meaningful patterns representative of the person s daily living habits. Any deviation from this learned profile is then considered as an unexpected situation. The study concerns on one hand the conception of a multivariate simulation process for generating large amount of data relevant to our experimental context. On the other hand, we build a generic unsupervised learning process for mining heterogeneous multivariate time-series and identifying temporal patterns. The proposed approach is applied in the context of identifying recurrent behaviors during activities of daily living. We especially evaluate the sensibility (tolerance of normal changes in behavior) and specificity (rejection of abnormal changes) of the system. This learning method is applied to the output of the simulation process and shows its relevance in extracting recurrents behaviors that can be a posteriori interpreted as some activities of daily living.
Complete list of metadata
Contributor : Florence Duchene <>
Submitted on : Wednesday, March 16, 2005 - 2:19:05 PM
Last modification on : Thursday, November 19, 2020 - 3:56:09 PM
Long-term archiving on: : Monday, September 20, 2010 - 12:21:13 PM


  • HAL Id : tel-00007607, version 2



Florence Duchêne. Fusion de Données Multicapteurs pour un Système de Télésurveillance Médicale de Personnes à Domicile. Interface homme-machine [cs.HC]. Université Joseph-Fourier - Grenoble I, 2004. Français. ⟨tel-00007607v2⟩



Record views


Files downloads