Search

As members of an increasingly aging society, one of our major priorities is to develop tools to detect the earliest stage of age-related disorders such as Alzheimer's Disease (AD). The goal of this paper is to evaluate the possibility of using unobtrusively collected activity-aware smart home behavior data to detect the multimodal symptoms that are often found to be impaired in AD. After gathering longitudinal smart home data for 29 older adults over an average duration of $>$ 2 years, we automatically labeled the data with corresponding activity classes and extracted time-series statistics containing ten behavioral features. Mobility, cognition, and mood were evaluated every six months. Using these data, we created regression models to predict symptoms as measured by the tests and a feature selection analysis was performed. Classification models were built to detect reliable absolute changes in the scores predicting symptoms and SmoteBOOST and wRACOG algorithms were used to overcome class imbalance where needed. Results show that all mobility, cognition, and depression symptoms can be predicted from activity-aware smart home data. Similarly, these data can be effectively used to predict reliable changes in mobility and memory skills. Results also suggest that not all behavioral features contribute equally to the prediction of every symptom. Future work therefore can improve model sensitivity by including additional longitudinal data and by further improving strategies to extract relevant features and address class imbalance. The results presented herein contribute toward the development of an early change detection system based on smart home technology.

International audience; Surface integrity directly affects the mechanical behavior of the workpiece, which is especially relevant on fatigue behavior. To characterize the quality of the machined surface, aspects such as material damage, roughness or residual stress are considered. Measurement of the material damage of the surface is characterized in some cases as surface drag, depth of the affected machining zone, a phenomenon which takes place due to plastic strain in the surface layer caused by machining stress which could have an influence on residual stress. Surface drag measurement done with optical microscopes has relevant uncertainty. In this paper, a methodology to measure the surface drag with lower uncertainty is proposed. The method consists of measuring the deformation of a grid as a result of the machining process. The grid was created with micromilling. The method was applied to analyze the effect of feed on the surface integrity after orthogonal cutting of Ti-6Al-4V. The depth of the affected layer was measured using a 3D optical measuring device (Alicona Infinite Focus IFG4) and compared with numerical simulations and a good agreement was achieved. In comparison with optical microscope results, it can be concluded that traditional method underestimates surface drag

Thermodiffusion, which leads to a component separation in a mixture due to the thermal gradient, still does not have an unambiguous microscopic picture. Therefore, experimental studies, especially in convection free environment, are important. As part of the 4th campaign on the DCMIX project, thermodiffusion experiments on three compositions of the toluene-methanol-cyclohexane ternary mixture, on a mixture of fullerene-tetralin-toluene and on a mixture of polystyrene-toluene-cyclohexane have been performed in microgravity conditions on board the International Space Station. A binary mixture of polystyrene-toluene has been filled into the companion cell for the campaign. The Selectable Optical Diagnostics Instrument (SODI), wich is a two-wavelength Mach-Zehnder interferometer for the ternary mixtures, plus a monochromatic Mach-Zehnder interferometer for the binary mixture, has been used in order to obtain the temperature and the concentration fields in the cells. Precisely, it is a 5-steps phase-shifting interferometry technique which is implemented with SODI, producing by means of laser illumination a set of 5 phase-shifted images of π/2 between them as function of the time. A total of 58 runs of various durations and at different mean temperatures have been conducted. Here, we evaluate the contrast of the interferograms, the quality in the phase stepping, the stability of the thermal regulation of the experiments and the level of environmental disturbances on board the space station during the campaign., SCOPUS: ar.j, info:eu-repo/semantics/published

International audience; Superconductor/Ferromagnet (S/F) hybrid systems show interesting magneto-transport behaviors that result from the transfer of properties between both constituents. For instance, magnetic memory can be transferred from the F into the S through the pinning of superconducting vortices by the ferromagnetic textures. The ability to tailor this type of induced behavior is important to broaden its range of application. Here we show that engineering the F magnetization reversal allows tuning the strength of the vortex pinning (and memory) effects, as well as the field range in which they appear. This is done by using magnetic multilayers in which Co thin films are combined with different heavy metals (Ru, Ir, Pt). By choosing the materials, thicknesses, and stacking order of the layers, we can design the characteristic domain size and morphology, from out-of-plane magnetized stripe domains to much smaller magnetic skyrmions. These changes strongly affect the magneto-transport properties. The underlying mechanisms are identified by comparing the experimental results to a magnetic pinning model. * javier.villegas@cnrs-thales.fr 2

This paper describes the European Space Agency (ESA) experiments devoted to study thermodiffusion of fluid mixtures in microgravity environment, where sedimentation and convection do not affect the mass flow induced by the Soret effect. First, the experiments performed on binary mixtures in the IVIDIL and GRADFLEX experiments are described. Then, further experiments on ternary mixtures and complex fluids performed in DCMIX and planned to be performed in the context of the NEUF-DIX project are presented. Finally, multi-component mixtures studied in the SCCO project are detailed. Graphical abstract: [Figure not available: see fulltext.]., 0, SCOPUS: re.j, info:eu-repo/semantics/published

We report on thermodiffusion experiments conducted on the International Space Station ISS during fall 2016. These experiments are part of the DCMIX (Diffusion and thermodiffusion Coefficients Measurements in ternary Mixtures) project, which aims at establishing a reliable data base of non-isothermal transport coefficients for selected ternary liquid mixtures. The third campaign, DCMIX3, focuses on aqueous systems with water/ethanol/triethylene glycol as an example, where sign changes of the Soret coefficient have already been reported for certain binary subsystems. Investigations have been carried out with the SODI (Selectable Optical Diagnostics Instrument) instrument, a Mach-Zehnder interferometer set up inside the Microgravity Science Glovebox in the Destiny Module of the ISS. Concentration changes within the liquids have been monitored in response to an external temperature gradient using phase-stepping interferometry. The complete data set has been made available in spring 2017. Due to additionally available measurement time, it was possible to collect a complete data set at 30∘C and an almost complete data set at 25∘C, which significantly exceeds the originally envisaged measurements at a single temperature only. All samples could be measured successfully. The SODI instrument and the DCMIX experiments have proven reliable and robust, allowing to extract meaningful data even in case of unforeseen laser instabilities. First assessments of the data quality have revealed six out of 31 runs with some problems in image contrast and/or phase step stability that will require more sophisticated algorithms. This publication documents all relevant parameters of the conducted experiments and also events that might have an influence on the final results. The compiled information is intended to serve as a starting point for all following data evaluations.

International audience; The storage modulus variation of anisotropic magnetorheological elastomers (MREs) induced by an external magnetic field was modelled in the frequency domain. This involves synthesising five anisotropic MREs with different particle content and measuring its dynamic and magnetic properties. Dynamic properties were measured using a rheometer equipped with a magnetorheological cell and the magnetic permeability of each sample was measured with a vibrating sample magnetometer. Scanning Electron Microscope images were used to determine particle distribution. A four parameter fractional derivative model was used to describe MRE viscoelasticity in the absence of magnetic field and the fitting error was not larger than 1%. Magneto-induced modulus was also studied and two different models were analysed, the one of Jolly et al. (Smart Mater Struct;5:607 (1999)) and the other one of López-López et al. (J Rheol. 56:1209 (2012)). The first model underestimated the influence of the magnetic field for low particle contents while at high ones it overestimated the magnetic field effect, up to 13%. However, in the second model magnetic permeability values were used, and the error between the model prediction and experimental data did not exceed 7%. Hence, a new linear magneto-viscoelastic model was proposed in frequency domain for anisotropic MREs based on López-López et al. model, which predicts the effect of magnetic field on the dynamic shear modulus in function of particle content and frequency.

International audience; It was designed and constructed a new thermogravitational column able to operate at high pressures (up to 50 MPa). This new thermogravitational column is of the cylindrical type with closed ends. It is made of stainless steel. The length of the column is 0.5 m and the gap between its two walls is variable. First, the column was validated at atmospheric pressure by means of measurements of the thermodiffusion coefficient of well-known binary mixtures. Then, this new thermogravitational column was used to measure the thermodiffusion coefficient of the binary mixtures 1,2,3,4-tetrahydronaphtaleneisobutylbenzene, 1,2,3,4-tetrahydronaphtalenen-dodecane, and isobutylbenzenen-dodecane at high pressures and within the pressure range between 0.1 and 20 MPa at a mean temperature of 25 °C. We have found a linear dependence between the thermodiffusion coefficient and the pressure.

The sorption isotherm, the glass transition, and the mutual diffusion coefficient of polyacrylamide/water solutions are obtained experimentally. All of these parameters are measured in the concentrated regime by gravimetric experiments. The mutual diffusion coefficient is also measured at high solvent concentrations by the sliding symmetric tubes technique. Three different polyacrylamide batches differing in their molar mass have been characterized. The results are expressed in terms of simple empirical correlations, suitable for use in process modelization or numerical simulations.

International audience; This contribution to materials selection theory aims at developing methodological methods and tools to analyse a complex set of material requirements with the objective of forecast whether there are materials that can fulfil it, or alternatively, if multi-materials selection is more likely to provide a solution. This "pre-analysis" of requirements examines the two main reasons which may prevent a single material solution: nonuniformity of materials space filling, or intrinsic contradiction between properties. A variety of statistical tools is used, based on Multivariate Analysis Methods as Principal Component Analysis (PCA) and the estimation of density distribution in the materials space. These tools allow to evaluate the "statistical compatibility" between the requirements and the available materials, and provides an estimate of the likelihood to find a single material solution or not. The methodology is applied to the research of a material for a machine tool frame.

International audience; After translation, the first stages traditionally involved in the materials selection are filtration and classification, which require formulation of criteria (constraints or objectives) deduced from information written in the set Of material requirements. These criteria, which are representative of the behavior of the material and the studied structure, must be formulated analytically in order to be used during selection stage. However, for complex behavior, analytical processing of models is no longer possible and it can be replaced by a combination of numerical resolution methods and an optimization method which make it possible to obtain approximate formal expressions of the criteria. In this paper, a complete selection method is proposed. The method is applied to the constraints as well as the objectives, in order to carry out the filtration and classification stages at the same time. The study of the thermomechanical behavior of a machine tool frame has been used to demonstrate the validity of the proposed method.