21549 research outputs found
Sort by
Étude expérimentale rétrospective de l'utilisation de la phényléphrine (néosynéphrine) dans un contexte chirurgical ophtalmologique
Get PDFDans un contexte chirurgical ophtalmologique, la phényléphrine par voie locale (Instillation oculaire de gouttes) est un médicament couramment employé afin d’obtenir une mydriase peropératoire. Le but de cette étude expérimentale rétrospective est d’évaluer objectivement la nature et l’intensité des effets non recherchés notamment cardiovasculaires de l’administration locale de phényléphrine, à partir des dossiers médicaux de chiens anesthésiés pour une chirurgie ophtalmologique de Janvier 2018 à Mars 2021 au Centre Hospitalier Universitaire Vétérinaire de l’École Nationale Vétérinaire de Toulouse. Les effets hypertenseur et bradycardisant de la néosynéphrine décrits dans la littérature se retrouvent sur les deux populations de chiens de cette étude, à savoir les animaux prémédiqués avec de la médétomidine, un ⍺2- agoniste (n=6) et ceux ayant reçu de l’acépromazine (n=11). Aucun effet sur la saturation pulsée de l’hémoglobine en oxygène n’a cependant été observé. La problématique du passage de la néosynéphrine dans la circulation systémique est peu étudiée dans la littérature et ne peut être caractérisée dans ce travail. Cependant un essai de caractérisation de la relation dose effet en fonction du nombre de gouttes reçues et du poids de l’animal a été réalisé sans résultat significatif. La faible taille des deux groupes étudiés ici est à prendre en compte dans cette analyse
Reynolds and mach number effects on the flow in a low-pressure turbine
No full textThis paper presents the Large Eddy Simulation (LES) of
a Low-Pressure Turbine (LPT) Nozzle Guide Vane (NGV) for
different Reynolds (Re) and Mach number (Ma). The analysis
is based on a slice of the blade that may be representative of
midspan flow where secondary flows, hub and shroud contributions
are lower. In LPT, the variation of the Re during the mission
of the gas turbine is a well-known effect since its value can vary
of a factor four between take-off and cruise. This can induce performance
variations due to various phenomena with among them
suction side boundary layer separation on the aft portion of the
blade due to an adverse pressure gradient and laminar boundary
layer that can be maintained due to the relatively low Re in
LPT. Similarly, the Ma in the LPT may vary depending on the
thrust required from the gas turbine at the considered mission
phase. The current paper investigates through numerical simulation
the flow representative of a medium-sized LPT with three
different Reynolds number Re = 175’000 (cruise), 280’000 (midlevel
altitude) and 500’000 (take-off) keeping the same characteristic
Mach number Ma = 0.2 and three different Mach number
Ma = 0.2, 0.5 and 0.8 keeping the same Reynolds number
Re = 280’000. The study focuses on different flow characteristics:
pressure distribution around the blade, near-wall flow behavior
and wake analysis. This includes the related generation
of losses and the effect of Re and Ma on these different phenomena.
A special emphasis is given to the generation of loss based
on an entropy approach and the redistribution of mean kinetic
energy towards turbulent kinetic energy. The results show that the increase of the Re has a destabilizing effect on potential separation
while the increase of the Ma has a stabilizing effect. The
peak in the TKE downstream of the blade is also moved upstream
closer to the trailing edge when the Ma is increased
Caractéristiques de l'élevage canin et félin en France : série d'enquêtes auprès des éleveurs.
Get PDFL’étude s’intéresse à l’élevage canin et félin en France à travers un recueil de données d’élevage. Après une mise en contexte de l’élevage français, les résultats d’une série de questionnaires réalisée auprès d’éleveurs canins et félins sont présentés dans le but de comparer ces deux types d’élevage. Il apparait des différences dans la localisation des élevages, la gestion des locaux, le temps de travail, le personnel ou encore les finances générées. D’autre part, les pratiques diffèrent en termes d’effectif d’animaux, d’utilisation des tests génétiques, de gestion de la reproduction, de socialisation des portées, de prophylaxie, de recours aux médecines alternatives, de biosécurité, de gestion de la vente et du relationnel avec les futurs acquéreurs
Pyrolysis of waste polyethylene in a semi-batch reactor to produce liquid fuel : optimization of operating conditions
Get PDFThis study investigated the interactive effects of temperature, residence time, and carrier gas flow rate on the liquid fuel production through the pyrolysis of waste polyethylene (WPE) in a bench-scale semi-batch reactor. To enhance the liquid fuel production, fifteen experiments were conducted based on a central composite design. The adaptive neural fuzzy model was adopted to establish the relationship between liquid fuel production and operating conditions. The R-squared value of the experimental and adaptive neural fuzzy model predicted that liquid fuel production was 0.9934. Four additional experimental results verified the adaptive neural fuzzy model's applicability. Subsequently, the genetic algorithm (GA) was adopted to optimize operating conditions to maximize liquid fuel production. The GA optimized operating conditions (temperature, residence time, and carrier gas flow rate) were: 488 °C, 20 min, and 20 mL/min. The liquid fuel under the optimal operating conditions was analyzed by Fourier-transform infrared spectroscopy (FTIR) and gas chromatography-mass spectrometry (GC–MS). The liquid fuel had similar main functional groups as diesel. The components of the liquid fuel were mainly 1-alkenes and n-alkanes ranging from C7 to C36. The effects of operating conditions on liquid fuel fractions and mean molecular weight were also investigated
A variational marginalized particle filter for jump Markov nonlinear systems with unknown transition probabilities
Get PDFThis paper studies a new variational marginalized particle filter for jointly estimating the state and the system mode parameters of jump Markov nonlinear systems. Contrary to the Markovian assumption usu- ally considered to model the evolution of the system modes, we introduce conjugate prior distributions for the system mode parameters. The joint posterior distribution of the state and system mode parame- ters is then marginalized with respect to the mode variables. The remaining state vector is sampled us- ing a sequential Monte Carlo algorithm, and the mode parameters are sampled using variational Bayesian inference. In order to obtain analytical solutions for the different variational distributions, we use an extended factorized approximation simplifying the variational distributions. A comprehensive simulation study is conducted to compare the performance of the proposed approach with the state-of-the-art for a modified nonlinear benchmark model and maneuvering target tracking scenarios
Vers l’application de l’apprentissage par renforcement inverse aux réseaux naturels d’attention
Get PDFLe cerveau humain, pour allouer de manière optimale les ressources attentionnelles limitées dont il dispose, supprime ou renforce l’activation de circuits neuronaux : il implémente des heuristiques. Dans une approche novatrice, nous proposons d’utiliser l’apprentissage par renforcement inverse pour caractériser la dynamique d’activation de ces réseaux. Un protocole expérimental est proposé, et les données collectées devraient permettre, à terme, de vérifier cette démarche
Correct-by-Construction Design of Hybrid Systems Based on Refinement and Proof
Get PDFHybrid systems are a wide category of systems consisting of multiple computers, linked together by some means and interacting with physical devices. Usually cheap, adaptable and incredibly versatile, these systems rapidly found their place in our everyday life, and can be found in basic devices like smart light bulbs, or in more complex structures such as smart factories or autonomous vehicles. In particular, due to their numerous advantages, they can nowadays be found in safety-critical domains, including avionics, self-driving cars or energy distribution. The complex and hybrid nature of hybrid systems make them quite hard to design in a safe way: while computer science is mostly able to deal with their discrete parts and control theory with their continuous part, we lack some kind of formalism that would be able to deal with both of these aspects and their integration with one another. The formal modelling of safety-critical hybrid systems is thus a major challenge of the domain. To overcome this challenge, we propose a generic formal framework, aimed at safely modelling hybrid systems in a correct-by-construction fashion, inherited from the Event-B method the framework is based on. This framework takes the form of an extensive set of theories that expands Event-B with mathematical features necessary to model continuous behaviours (e.g. continuous functions and differential equations), a generic model that encodes a generic hybrid system, complete with its controller and continuous plant, and a series of patterns, based on refinement, that allow to ease the design process. In particular, we defined three formal patterns, inspired by general practice in hybrid system designs: approximation (substituting an equation system with an approximately equivalent one), centralised control with multiple plants (splitting a plant into a set of plant with a centralised controller) and distributed hybrid systems (system made of components that each consist of a controller and a plant, enforcing a global invariant). This framework is designed to be extensible: adding a new component is done by describing it as a refinement of the generic model, and possibly to accompany it with relevant theories. We used our framework on various problems taken from control theory, including computer-assisted cars, water tanks, robots and inverte
Improving user experience of SSVEP-BCI through reduction of stimuli amplitude depth
No full textSteady-States Visually Evoked Potentials (SSVEP)
is currently one of the most widely used paradigms in
Brain-Computer Interfaces (BCI). Although the SSVEP-BCI
are characterized by their high and robust classification
performance, the repetitive presentation of flickering stimuli is
uncomfortable from a user experience perspective point of view.
Indeed, the low-level visual features of SSVEP stimuli make
them straining to the eyes over time and could be disruptive to
the execution of tasks requiring sustained attention. They could
even induce epileptic seizures. This study explores the reduction
of the stimulation amplitude depth (a magnitude diminution of
90%) for the design of SSVEP stimuli as a solution to improve
user comfort. The classification accuracy obtained by different
pipelines was systematically compared between low amplitude
and standard full amplitude SSVEP stimuli. The results reveal
high classification accuracy for both high (99.8%) and low
magnitude (80.2%) stimuli using the Task-Related Component
Analysis (TRCA) classification method. The present findings
demonstrate the validity of reducing SSVEP stimuli amplitude
to increase users’ comfort paving the way for transparent BCI
operation
Crossover between Re-Nucleation and Dendritic Growth in Electrodeposition without Supporting Electrolyte
Get PDFThis work deals with the formation of dendritic structures by electrodeposition of Cu2+ and Ag+ without supporting electrolyte in Hele-Shaw cells. The transition between the two main patterns, ramified branches and dendrites, is specifically addressed at the scale of branch microstructure using careful SEM observations. Ramified branches, composed only of grain assemblies, are obtained at low current densities because of a re-nucleation process induced by space charge dynamics (Fleury, Nature, 1997). For current densities higher than a given threshold, ramified branches are also formed by re-nucleation but another growth mode, the dendritic growth, is also observed while, at the macro-scale, the pattern remains fractal and isotropic. This shows that 1) pattern transition originates from a morphological transition at microstructure scale and 2) the re-nucleation process enables a freedom in local growth direction allowing the pattern to be fractal at the macro-scale. The onset of the dendritic growth mode, from shape instability of the grains, is considered with Mullins & Sekerka model. This latter disagrees with the observations by predicting that the grains are always unstable. It is proposed that the space charge plays a key role by controlling the shape stability and thus the transition between the two growth modes
Évaluation de la fonction cognitive postanesthésique du chien et du chat sains.
Get PDFQu’ils soient transitoires ou permanents, les troubles et déficits cognitifs faisant suite à une anesthésie générale sont maintenant bien décrits chez l’homme. Malgré un recours assez fréquent à l’anesthésie générale chez les animaux de compagnie, peu de choses dans ce domaine sont actuellement décrites chez les carnivores domestiques. Le but de cette étude expérimentale est donc d’évaluer les éventuels troubles cognitifs postanesthésiques chez des chiens et des chats sains, jeunes et en bonne santé suite à une stérilisation chirurgicale dite de convenance. Pour se faire, un questionnaire en trois parties (préanesthésique, postanesthésique à 48 heures et à 15 jours) a été distribué aux propriétaires afin d’évaluer le comportement de leur animal. Suite à l’analyse descriptive des résultats, l’étude ne montre a priori pas l’existence de déficits cognitifs cliniquement handicapants suite à une anesthésie générale chez le chien et le chat sain, après une intervention chirurgicale de stérilisation