HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
Metabolism, application in the food industry, and enrichment strategies of gamma-aminobutyric acid
No full textInternational audienceBackgroundGamma-aminobutyric acid (GABA) is a non-protein amino acid with multiple physiological functions and health benefits that occurs naturally in various organisms. GABA has a broad application prospect in food industry.Scope and approachThe metabolism and application in food preservation and processing of GABA, as well as processing methods for increasing its content were reviewed.Key findings and conclusionsGABA metabolism is involved in the tricarboxylic acid (TCA) cycle and can affect energy and substance metabolism. GABA has shown a positive role in food preservation, particularly in alleviating the cold damage, browning, softening, senescence and oxidative damage of food, and improving pathogen resistance and nutritional quality. This is mainly due to the self-decomposition of exogenous GABA and the synthesis and decomposition of endogenous GABA activated by exogenous GABA, which regulate the complex metabolic activities of food via the TCA cycle. GABA as a food constituent enables the development of functional foods with health benefits. Additionally, the appropriate dose of GABA can also improve the functional properties of proteins, thus positively impacting on physicochemical properties of foods, which can facilitate the development of functional food formulations. Germination treatments, microbial fermentation, and innovative physical processing technologies are the most popular methods for increasing GABA levels in foods. Microbial fermentation can enrich GABA in foods that do not contain it. Innovative physical processing technologies can enhance not only the synthesis of GABA, but also the levels of other nutrients. The combination of different processing methods is the trend of GABA enrichment
Metal or Metal-Free Catalysis for the Oxidation of Biosourced Furfural
No full textInternational audienceGreen chemistry and sustainable development are becoming increasingly indispensable for finding solutions to climate change, organizing a local circular economy, ensuring energy security, and promoting autonomy based on biosourced raw materials. Lignocellulosic biomass is one of the essential raw materials to produce biofuels and biosourced chemicals that can partially replace fossil resources. The development of conversion technologies for these biosourced products is growing rapidly. Furfural plays a major role as an industrial product derived from this renewable material, sparking interest in efficiently and cleanly transforming it. In this review, we present oxidation transformation methods of furfural into resulting molecules such as furanones, C4 acids, and furoic acid. This is done based on different metallic and metal-free catalytic systems to allow readers to compare their advantages and disadvantages in the perspective of sustainable development
Les produits de contraste en imagerie médicale : usages, caractéristiques, risques et remboursement
No full textThis report delves into the utilization of Contrast Agents in France, providing a thorough exploration of their crucial role in medical imaging. It offers a detailed examination of their composition, regulatory framework, associated risks, and economic implications. Additionally, the report analyzes various medical applications, such as intravenous urography and retrograde cystography, highlighting their significance in diagnosing diseases affecting the urogenital system, amongs others.The regulatory framework of contrast agents is elucidated, with a particular focus on their classification, as they are considered medicaments. These agents are subject to control and supervision by the National Agency for Drug and Health Product Safety (ANSM). The report outlines adverse effects and risks, including allergic reactions and kidney dysfunction, along with preventive measures.To cover diverse range of aspects of Contrast Agents, extensive research was conducted within the current health policy in France. The report underscores the functioning of the patient supply chain, starting from the approval of marketing authorization (AMM) and extending to surveillance, costs, and reimbursements. These costs, linked to contrast agents, are being reviwed by a reform to integrate them into the existing patient supply chain, aiming to benefit all stakeholders.This report provides a comprehensive overview, allowing the identification of how the contrast agents market operates in France throughout its life cycle.Ce Mémoire d’Intelligence Méthodologique plonge en profondeur dans l'utilisation des produits de contraste en France, offrant une exploration approfondie de leur rôle essentiel en imagerie médicale, de leur composition, des enjeux réglementaires, des risques associés, et des implications économiques. L'étude analyse les divers modalités d'imagerie, tels que l'urographie intraveineuse et la cystographie rétrograde, mettant en avant leur importance cruciale dans le diagnostic des affections urogénitales, notamment le reflux vésico-urétéral traité à travers la cystographie rétrograde. Une attention particulière est portée aux diverses applications cliniques. Le cadre réglementaire encadrant les produits de contraste est décrypté, soulignant leur classification en tant que médicaments et les strictes régulations de l'Agence nationale de sécurité du médicament et des produits de santé. Les risques, tels que les réactions allergiques, l'insuffisance rénale, et les interactions médicamenteuses, sont analysés en détail, présentant une gestion proactive des risques qui souligne l'importance d'une évaluation préalable de la fonction rénale et d'une prévention des réactions allergiques. La section sur le remboursement et la politique de santé en France met en lumière le processus de demande d'autorisation de mise sur le marché, la pharmacovigilance, et explique le remboursement des produits de contraste dans le cadre du forfait technique. Les coûts associés aux produits de contraste sont également examinés, avec des détails sur la récente réforme modifiant le circuit d'approvisionnement. Ainsi, ce mémoire offre une vision exhaustive, recensant la situation des produits de contraste en France, embrassant le cadre réglementaire, et soulignant la nécessité impérative d'une gestion rigoureuse des risques
Context-awareness in Pedagogical Resources Recommender System for Teachers
No full textInternational audienceIn today's fast-paced world, teachers seek continuous improvement for their courses to match their context using e-learning systems. With the aid of technologies, teachers have access to vast amount of resources which requires guidance in selecting appropriate pedagogical resources according to their context. Pedagogical resources recommender system should take into account this context to offer best recommendations dedicated to this user-teacher's context. This paper explores the impact of context-awareness in enhancing the performance of pedagogical resources recommender systems, focusing on three main approaches: contextual pre-filtering, contextual post-filtering, and contextual modeling. This paper proposes an enhancement to the context-awareness of a 2D pedagogical resources recommender system by integrating the contextual modeling approach into a collaborative-filtering recommendation technique. The evaluation of the proposed approach showed a significant improvement of recommendation accuracy compared to the contextual prefiltering and post-filtering approaches
Copula-based conformal prediction for object detection: a more efficient approach
No full textInternational audienceObject detection is an important vision task, and providing statistical guarantees around such detections can be of critical importance. So far, most conformal bounding box regression approaches do not simultaneously account for heteroscedasticity and dependencies between the residuals of each dimension. In this paper, we examine the importance of such dependencies and heteroscedasticity in the context of multi-target conformal regression, we apply copula-based conformal prediction methods to model them and to improve the volume of bounding box prediction regions. We compare these methods to the state-of-the-art conformal object detection approaches (on the KITTI and the BDD100K autonomous driving benchmarks) and the empirical copula-based method shows high-efficiency results that are robust w.r.t. heteroscedasticity and also robust w.r.t. the structure of the dependencies
Reconstruction 3D des voies biliaires pour l’amélioration de la CholangioPancréatographie Rétrograde par voie Endoscopique (CPRE)
No full textThis thesis addresses the challenge of improving Endoscopic Retrograde Cholangiopancreatography (ERCP) by proposing an advanced method for 3D reconstruction of bile ducts. Current ERCP techniques face significant limitations, mainly due to the difficulty of visualizing bile ducts using 2D images, which can lead to planning errors and complications during interventions. The use of 3D reconstruction of bile ducts from MR cholangiography (MRCP) images presents a promising solution to overcome these challenges. The thesis introduces a novel approach for automatic segmentation called BDU-Net, based on the U-Net architecture and specifically tailored to MRCP images. The performance of BDU-Net was compared with other reference segmentation models, such as V-Net and U-Net++, using quantitative metrics like the Dice score and Hausdorff distance, as well as a qualitative assessment by experts. The results show a significant improvement in the accuracy of 3D reconstructions with BDU-Net. Additionally, the thesis explores the impact of various preprocessing methods for MRCP images, including dynamic cropping, to enhance the segmentation quality. This preprocessing step improves the visualization of bile duct structures and enhances 3D reconstruction, especially in complex cases. Beyond segmentation contributions, the thesis also presents a plugin called CPRE Pro, developed for the 3DSlicer platform. This tool integrates 3D reconstructions into the clinical workflow, facilitating ERCP planning. With an intuitive interface, clinicians can interact with MRCP images, perform automatic segmentations, and visualize bile duct anatomy in 3D, thus improving clinical decision-making and intervention planning.Cette thèse aborde la problématique de l'amélioration de la CholangioPancréatographie Rétrograde par voie Endoscopique (CPRE), en proposant une méthode avancée de segmentation et de reconstruction 3D des voies biliaires. Les techniques actuelles de la CPRE présentent des limites importantes, notamment en raison de la difficulté de la visualisation des voies biliaires à partir des images 2D, ce qui peut entraîner des erreurs de planification et des complications durant les interventions. La reconstruction 3D des voies biliaires, à partir des images de cholangio-IRM, apparaît comme une solution prometteuse pour surmonter ces difficultés. La thèse propose une nouvelle approche de segmentation automatique BDU-Net, à base du réseau U-Net, spécifiquement adaptée aux caractéristiques des images de cholangio-IRM. Les performances du BDU-Net ont été comparées avec celles de réseaux de segmentation de références, tels que V-Net et U-Net++, en utilisant des métriques quantitatives comme le score de Dice et la distance de Hausdorff, ainsi qu'une évaluation qualitative réalisée par des experts. Les résultats montrent une amélioration de la précision des reconstructions 3D grâce à BDU-Net. Cette thèse explore également l'impact des différentes méthodes de prétraitement des images de cholangio-IRM, incluant l'utilisation du recadrage dynamique, pour améliorer la qualité de segmentation. Ce prétraitement permet de mieux visualiser les structures biliaires, et d'améliorer la reconstruction 3D des voies biliaires, particulièrement dans les cas complexes. En plus des contributions en segmentation, cette thèse propose un plugin, nommé CPRE Pro, développé pour la plateforme 3DSlicer. Cet outil permet d'intégrer les reconstructions 3D dans le workflow clinique, facilitant la planification de la CPRE. Grâce à une interface intuitive, les cliniciens peuvent interagir avec les images de cholangio-IRM, effectuer des segmentations automatiques et visualiser l'anatomie biliaire en 3D
From Individual to Group: Developing a Context-Aware Multi-Criteria Group Recommender System
No full textInternational audienceGroup decision-making is becoming increasingly common in areas such as education, dining, travel, and finance, where collaborative choices must balance diverse individual preferences. While conventional recommender systems are effective in personalization, they fall short in group settings due to their inability to manage conflicting preferences, contextual factors, and multiple evaluation criteria. This study presents the development of a Context-Aware Multi-Criteria Group Recommender System (CA-MCGRS) designed to address these challenges by integrating contextual factors and multiple criteria to enhance recommendation accuracy. By leveraging a Multi-Head Attention mechanism, our model dynamically weighs the importance of different features. Experiments conducted on an educational dataset with varied ratings and contextual variables demonstrate that CA-MCGRS consistently outperforms other approaches across four scenarios. Our findings underscore the importance of incorporating context and multi-criteria evaluations to improve group recommendations, offering valuable insights for developing more effective group recommender systems
Étude expérimentale et numérique de l’endommagement et de la rupture des capsules en écoulement
No full textCapsules, composed of fluid droplets surrounded by an elastic membrane, play an essential role in protecting and transporting active ingredients. However, they can deformor rupture under the influence of hydrodynamic stresses, necessitating precise analysis of their flow behavior in order to control the release of encapsulated substances. The aim of this study is therefore to study the damage and rupture of capsules in flow, using both experimental and numerical approaches. A microrheometric study was carried out to examine these phenomena on ovalbumin capsules. An inverse analysis method was used to determine the surface shear modulus of the deformed capsules. The temporal evolution of this modulus shows a decrease, directly linked to the capsule damage. Rupture is characterized by comparing the number and size of capsules on a 10-μm-sample before and after shearing, showing an increase in the percentage of ruptured capsules with their size, the exposure time to shear and the capillary number. In the numerical part of the study, a fluid-structure interaction model is employed to simulate the behavior of a capsule in hyperbolic flow, accounting for membrane damage and rupture. The results are then compared with existing experimental data. Our numerical analysis reveals that the description of rupture localization is influenced by the presence of compressive zones. To overcome this restriction, we include a bending resistance in the numerical model, aiming to regulate the formation of wrinkles on the membrane.Les capsules, composées de gouttelettes fluides entourées d’une membrane élastique, jouent un rôle essentiel dans la protection et le transport des principes actifs. Toutefois, elles peuvent se déformer ou se rompre sous l’influence des contraintes hydrodynamiques, nécessitant une analyse précise de leur comportement en écoulement afin de contrôler la libération des substances encapsulées. À cet égard, ces travaux visent à étudier l’endommagement et la rupture des capsules en écoulement, via des approches expérimentales et numériques. Une étude microrhéométrique est menée pour examiner ces phénomènes sur des capsules d’ovalbumine. Une méthode d’analyse inverse a permis de déterminer le module de cisaillement surfacique des capsules déformées. L’évolution temporelle de ce module montre une diminution, directement liée à l’endommagement des capsules sous l’effet du cisaillement. La rupture est caractérisée en comparant le nombre et la taille des capsules avant et après le cisaillement, montrant une augmentation du pourcentage de capsules rompues avec leur taille, la durée de cisaillement et le nombre capillaire. Dans la partie numérique, un modèle d’interaction fluide-structure est appliqué pour simuler le comportement d’une capsule en écoulement hyperbolique, tenant compte de l’endommagement et de la rupture de sa membrane. Les résultats sont ensuite comparés aux données expérimentales existantes. Notre analyse numérique révèle qu’en cas de rupture, la description de la localisation est affectée par la présence des zones compressives. Pour y pallier, une résistance à la flexion est intégrée dans le modèle numérique, visant à régulariser la formation des plis sur la membrane
Control and real-time experiments for a multi-agent aerial transportation system
No full textInternational audienceThis work presents a Multi-Agent Aerial Transportation System (MAATS) with various aerial robots carrying a cable suspended load. First, the dynamical model of the coupled system is described using the Newton-Euler formalism, considering a mass connected to the aerial drones by rigid cables. The objective is to control the load's position to perform trajectory tracking using n drones. Hence, a hierarchical controller is designed using Lyapunov-like energy functions, where the resultant desired cable tension is employed as a virtual control input for the load control which in turn is distributed among the agents to obtain a suitable desired pose of each agent that ensures the trajectory tracking of the load. Then, position and attitude control of each drone is carried out using the desired pose. The stability analysis of the closed-loop system is provided, demonstrating the stability of the coupled system. Finally, the performance of the MAATS control strategy is verified in software-in-the-loop simulations, as well as in real-time experiments, considering 2 and 3 drones, and showing good behavior in spite of significant external perturbations acting on the cables, the load or the agents
Uncertainty propagation in stereo matching using copulas
No full textInternational audienceThis contribution presents a concrete example of uncertainty propagation in a stereo matching pipeline. It considers the problem of matching pixels between pairs of images whose radiometry is uncertain and modeled by possibility distributions. Copulas serve as dependency models between variables and are used to propagate the imprecise models. The propagation steps are detailed in the simple case of the Sum of Absolute Difference cost function for didactic purposes. The method results in an imprecise matching cost curve. To reduce computation time, a sufficient condition for conserving possibility distributions after the propagation is also presented. Finally, results are compared with Monte Carlo simulations, indicating that the method produces envelopes capable of correctly estimating the matching cost