nam University

Saint-Sauveur-en-Rue, France

nam University

Saint-Sauveur-en-Rue, France
Time filter
Source Type

Guerreschi E.,nam University | Guerreschi E.,University of Notre Dame | Humeau-Heurtier A.,University of Notre Dame | Mahe G.,University of Angers | And 3 more authors.
Biomedical Signal Processing and Control | Year: 2013

Quantifying and modelling the cardiovascular system (CVS) represent a challenge to improve our understanding of the CVS. To describe and quantify the CVS, several physiological signals have been analyzed through various signal processing methods. Recently, a quantitative descriptor - the multiscale entropy (MSE) - has been proposed to quantify time series complexity (i.e. the degree of regularity of signal fluctuations) over multiple time scales. Heart rate variability (HRV) signals (i.e. data from the heart) have largely been studied through MSE. By contrast, complexities of signals from the macrocirculation (i.e. elastic and muscular arteries) and the microcirculation (i.e. arterioles and capillaries), two other main components of the CVS, have rarely been investigated simultaneously with MSE. We therefore propose to quantify and compare complexity of signals from these three CVS subsystems: the heart, the macrocirculation and the microcirculation, using MSE. Electrocardiograms, electrical bio-impedance signals (macrocirculation), as well as laser Doppler flowmetry (LDF) signals from finger and forearm (microcirculation) have been recorded simultaneously in nine healthy subjects. The MSE values from these data have been computed and compared. We note a significant lower complexity on scales τ = 1, 2 and 3 (i.e. around 1.08 Hz, 0.54 Hz and 0.36 Hz respectively) for LDF signals from the finger compared to the ones of signals from the heart and the macrocirculation. On scale τ = 5 (i.e. 0.21 Hz), complexity value of signals from the macrocirculation is significantly lower than the ones of HRV and data from the microvascular blood flow in forearm. The three CVS subsystems present different complexity values depending on scales. It could now be of interest to investigate if these complexity differences are due to physiological activities. Moreover, our results could be compared with those obtained from data recorded on patients with vascular diseases. © 2013 Elsevier Ltd. All rights reserved.

Glocheux Y.,Queen's University of Belfast | Albadarin A.B.,Queen's University of Belfast | Galan J.,Complutense University of Madrid | Oyedoh E.,Queen's University of Belfast | And 5 more authors.
Microporous and Mesoporous Materials | Year: 2014

This work presents the possibility of optimising 3D organised mesoporous silica (OMS) coated with both iron and aluminium oxides for the optimal removal of As(III) and As(V) from synthetic contaminated water. The materials developed were fully characterised and were tested for removing arsenic in batch experiments. The effect of total Al to Fe oxides coating on the selective removal of As(III) and As(V) was studied. It was shown that 8% metal coating was the optimal configuration for the coated OMS materials in removing arsenic. The effect of arsenic initial concentration and pH, kinetics and diffusion mechanisms was studied, modelled and discussed. It was shown that the advantage of an organised material over an un-structured sorbent was very limited in terms of kinetic and diffusion under the experimental conditions. It was shown that physisorption was the main adsorption process involved in As removal by the coated OMS. Maximum adsorption capacity of 55 mg As(V) g-1 was noticed at pH 5 for material coated with 8% Al oxides while 35 mg As(V) g -1 was removed at pH 4 for equivalent material coated with Fe oxides. © 2014 Elsevier Inc. All rights reserved.

Basha B.C.,University of Rennes 1 | Pillement S.,nam University | Piestrak S.J.,University of Lorraine
Proceedings - IEEE International Symposium on Circuits and Systems | Year: 2015

Field Programmable Gate Arrays (FPGAS) used in mission-critical applications such as aerospace, nuclear, and defense require high reliability in spite of internal faults. Fortunately, today's FPGAS have the ability to dynamically reconfigure themselves in the field, which may help to mitigate the effects of certain faults affecting the FPGA devices. Although the reconfiguration process can remove only the upsets affecting the configuration bitstream, unfortunately, there are other sources of faults that might directly affect hardware resources of reconfigurable FPGAS. Their nature and consequences differ from those which occur in the configuration bitstream and their effects cannot be corrected by performing configuration writeback. This paper proposes a fault-aware configurable logic block architecture to detect such faults in FPGA-implemented logic circuits. The fault coverage of the proposed architecture is also discussed. Hardware complexity estimations suggest higher efficiency of the approach proposed over similar existing ones. © 2015 IEEE.

Le Bastard C.,nam University | Baltazart V.,nam University | Derobert X.,nam University | Wang Y.,University of Nantes
2012 14th International Conference on Ground Penetrating Radar, GPR 2012 | Year: 2012

In the field of civil engineering, sounding the layers is classically performed using standard ground-penetrating radar (GPR), whose vertical resolution is bandwidth dependent. The layer thicknesses are deduced from both the time delays of backscattered echoes and the dielectric constants of the layers. In contrast with the conventional spectral analysis approaches, we propose in this paper to use one of the most powerful machine learning algorithm, namely the Support Vector Machine(SVM), to perform Time Delay Estimation (TDE) of backscattered radar signals. In particular, this paper demonstrates the super time resolution capability of such technique in the context of overlapping and totally correlated echoes when thin pavement layers survey is under scope. ©2012 IEEE.

Amhaz R.,nam University | Amhaz R.,National Polytechnic Institute of Toulouse | Chambon S.,National Polytechnic Institute of Toulouse | Idier J.,University of Nantes | Baltazart V.,nam University
2014 IEEE International Conference on Image Processing, ICIP 2014 | Year: 2014

This paper proposes a new algorithm for crack detection based on the selection of minimal paths. It takes account of both photometric and geometric characteristics and requires few information a priori. It is validated on synthetic and real images. © 2014 IEEE.

Loading nam University collaborators
Loading nam University collaborators