Time filter

Source Type

Grenoble, France

Fleury A.,Ecole Des Mines de Douai | Vacher M.,CNRS Informatics Laboratory of Grenoble | Noury N.,TIMC IMAG Laboratory | Noury N.,University Claude Bernard Lyon 1
IEEE Transactions on Information Technology in Biomedicine | Year: 2010

By 2050, about one third of the French population will be over 65. Our laboratorys current research focuses on the monitoring of elderly people at home, to detect a loss of autonomy as early as possible. Our aim is to quantify criteria such as the international activities of daily living (ADL) or the French Autonomie Gerontologie Groupes Iso-Ressources (AGGIR) scales, by automatically classifying the different ADL performed by the subject during the day. A Health Smart Home is used for this. Our Health Smart Home includes, in a real flat, infrared presence sensors (location), door contacts (to control the use of some facilities), temperature and hygrometry sensor in the bathroom, and microphones (sound classification and speech recognition). A wearable kinematic sensor also informs postural transitions (using pattern recognition) and walk periods (frequency analysis). This data collected from the various sensors are then used to classify each temporal frame into one of the ADL that was previously acquired (seven activities: hygiene, toilet use, eating, resting, sleeping, communication, and dressing/undressing). This is done using support vector machines. We performed a 1-h experimentation with 13 young and healthy subjects to determine the models of the different activities, and then we tested the classification algorithm (cross validation) with real data. © 2009 IEEE. Source

Hadidi T.,TIMC IMAG Laboratory | Noury N.,INSA Lyon
2011 IEEE 13th International Conference on e-Health Networking, Applications and Services, HEALTHCOM 2011 | Year: 2011

Modeling the behavior of a system from observation is the first step in understanding a phenomenon. The model then opens the way to simulation in order to reproduce the behavior of the real system with the flexibility to modify, adjust and play with various scenarios. In this paper, we present a simulator of the activity of elderly persons in a Health Smart Home using the HMM "Hidden Markov Model". This simulator aims at modeling the effects of the loss of autonomy for the elderly living independently. We test several correlation methods to evaluate the similarity between real and simulated data. The experimental data was obtained in the HIS "Habitat Intelligent pour la Santé" within the framework of the French national "AILISA" project. © 2011 IEEE. Source

Ali E.S.M.,Carleton University | Ali E.S.M.,Ottawa Hospital Cancer Center | Spencer B.,Carleton University | Spencer B.,TIMC IMAG Laboratory | And 2 more authors.
Physics in Medicine and Biology | Year: 2015

In this study, a quantitative estimate is derived for the uncertainty in the XCOM photon mass attenuation coefficients in the energy range of interest to external beam radiation therapy - i.e. 100keV (orthovoltage) to 25MeV - using direct comparisons of experimental data against Monte Carlo models and theoretical XCOM data. Two independent datasets are used. The first dataset is from our recent transmission measurements and the corresponding EGSnrc calculations (Ali et al 2012 Med. Phys. 39 5990-6003) for 10-30MV photon beams from the research linac at the National Research Council Canada. The attenuators are graphite and lead, with a total of 140 data points and an experimental uncertainty of∼0.5% (k=1). An optimum energy-independent cross section scaling factor that minimizes the discrepancies between measurements and calculations is used to deduce cross sectionuncertainty. The second dataset is from the aggregate of cross sectionmeasurements in the literature for graphite and lead (49 experiments, 288 data points). The dataset is compared to the sum of the XCOM data plus the IAEA photonuclear data. Again, an optimum energy-independent cross sectionscaling factor is used to deduce the cross sectionuncertainty. Using the average result from the two datasets, the energy-independent cross sectionuncertainty estimate is 0.5% (68% confidence) and 0.7% (95% confidence). The potential for energy-dependent errors is discussed. Photon cross sectionuncertainty is shown to be smaller than the current qualitative 'envelope of uncertainty' of the order of 1-2%, as given by Hubbell (1999 Phys. Med. Biol 44 R1-22). © 2015 Institute of Physics and Engineering in Medicine. Source

Gennai S.,Grenoble University Hospital Center | Gennai S.,TIMC IMAG Laboratory | Maignan M.,Grenoble University Hospital Center | Maignan M.,TIMC IMAG Laboratory | And 5 more authors.
Experimental Lung Research | Year: 2015

Objective: To evaluate the effects of 1 and 5 M of Cyclosporine A (CsA), administered 24 hours after a cold ischemic period, in an ex vivo reperfused pig lung model. Methods: The experiments were performed in 15 pigs. Each pair of lungs was surgically separated. Extracorporeal perfusion and mechanical ventilation were started after a cold ischemia of 2 hours for one lung and 24 hours for the contralateral. We constituted three groups (n = 5 each): two groups for which the lung underwent a 24-hour ischemia received either 1 or 5 M of CsA at the time of reperfusion, and a control group without CsA. For each group, lungs undergoing a 2-hour ischemia did not receive CsA. Results: Reperfusion with either CsA increased the PO2 levels in a dose dependent manner, and reduced concentrations of the receptor for advanced glycation endproducts, compared to the control. The pulmonary arterial pressure, the capillary pressure, and the pulmonary vascular resistances were not increased, even with 5 M of CsA. No significant change was shown on cytokines levels. Discussion: Postconditioning with CsA improves lung function, after a 24-hour cold ischemic period. Either 1 or 5 M seemed to be safe regarding the pulmonary vascular pressures and resistances. © 2015 Taylor & Francis Group, LLC. Source

Glade N.,TIMC IMAG Laboratory | Glade N.,AGIM Laboratory | Elena A.,AGIM Laboratory | Corblin F.,TIMC IMAG Laboratory | And 5 more authors.
Proceedings - 25th IEEE International Conference on Advanced Information Networking and Applications Workshops, WAINA 2011 | Year: 2011

This paper aims at warning modellers in systems biology against several traps encountered in the modelling of Boolean thresholded automata networks, i.e. the Hopfield-like networks that are often used in the context of neural and genetic networks. It introduces a new manner based on inverse methods to conceive such models. Using these techniques, we re-visit the model of regulatory network of Arabidopsis thaliana morphogenetic network. In this context, we discuss about the non-uniqueness of models, on a possible taxonomy of the set of valid models and on the sense of the relative size of the basin of attractions within or between these models. © 2011 IEEE. Source

Discover hidden collaborations