Towards non-intrusive sleep pattern recognition in elder assistive environment

Hongbo Ni (1, 2) , Bessam Abdulrazak (1) , Daqing Zhang (3, 4, 5) , Shu Wu (6) , Zhiwen Yu (2) , Xingshe Zhou (2) , Shengrui Wang (7)

1 DOMUS - DOmotique et informatique Mobile à l'Université de Sherbrooke
2 SCS-NPU - School of Computer Science - Northwestern Polytechnical University
3 TSP - RST - Département Réseaux et Services de Télécommunications
4 SAMOVAR - Services répartis, Architectures, MOdélisation, Validation, Administration des Réseaux
5 CNRS - Centre National de la Recherche Scientifique
6 UdeS - Université de Sherbrooke
7 MOIVRE Centre

Quality of sleep is an important attribute of an elder's health state and its assessment is still a challenge. The sleep pattern is a significant aspect to evaluate the quality of sleep, and how to recognize elder's sleep pattern is an important issue for elder-care community. With the pressure sensor matrix to monitor the elder's sleep behavior in bed, this paper presents an unobtrusive sleep postures detection and pattern recognition approaches. Based on the proposed sleep monitoring system, the processing methods of experimental data and the classification algorithms for sleep pattern recognition are also discussed.

Mots clés: Sleep pattern; Elder-care; Pressure sensor; Naïve bayes; Random forest

Domaines: Réseaux et télécommunications [cs.NI]; Santé

Hongbo Ni, Bessam Abdulrazak, Daqing Zhang, Shu Wu, Zhiwen Yu, et al.. Towards non-intrusive sleep pattern recognition in elder assistive environment. Journal of ambient intelligence and humanized computing, 2012, 3 (2), pp.167-175. ⟨hal-00751325⟩