HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
Optimisation de l'organisation du bloc opératoire au CHU de Donka (Guinée) - Élaboration d'un guide pratique
No full textThe operating theatre plays a central role in patient care, both for scheduled procedures and emergency surgery. An optimally organised operating environment directly contributes to care quality, patient safety and favourable working conditions for healthcare teams. Organisational practices have evolved through medical progress, technological advances and the development of new tools supporting planning and risk management.This project aims to analyze the organisation of operating theatres in France and to compare them with that of the operating room at Donka University Hospital in Guinea, in order to identify improvement strategies adapted to a low-resource hospital setting, locally referred to as a “zone jaune” (low-resource setting). The analysis highlights several key issues, including surgical scheduling, patient flow management, team coordination, management of logistics and medical devices and the implementation of a structured quality approach. Based on best practices observed in French models and the literature, a set of recommendations was developed to support the evolution of Donka’s operating theatre towards a more efficient, harmonised and safer organisation.Le bloc opératoire occupe une place essentielle dans la prise en charge des patients, que ce soit pour les interventions programmées ou pour la chirurgie d’urgence. Une organisation optimale de cet environnement contribue directement à la qualité des soins, à la sécurité des patients et aux conditions de travail des équipes soignantes. Les pratiques organisationnelles ont évolué grâce aux avancées médicales, aux progrès technologiques et au développement de nouveaux outils de planification et de gestion des risques. Ce mémoire a pour objectif d’étudier l’organisation de blocs opératoires en France et de la comparer à celle du bloc opératoire du Centre Hospitalier Universitaire (CHU) de Donka, en Guinée, afin d’identifier des axes d’amélioration adaptés à un contexte hospitalier en zone dite « jaune ». L’analyse met en évidence plusieurs enjeux, notamment la planification des interventions, la gestion du parcours patient, la coordination des équipes, la gestion de la logistique et des équipements médicaux et la mise en place d’une démarche qualité structurée. Sur la base des bonnes pratiques tirées des modèles français et de la littérature, des recommandations ont été formulées pour accompagner l’évolution du bloc opératoire de Donka vers une organisation plus efficace, harmonisée et sécurisée
Favoriser la chirurgie ambulatoire dans le parcours patient
No full textWith an ambulatory surgery rate of 64% in 2024, France remains below the recommended target of 80%. This thesis identifies the barriers (human, organizational, technical, financial) and facilitators to the development of ambulatory surgery. The study is based on an analysis of the French regulatory framework, institutional recommendations, international comparisons and field observations. Identified facilitators include Enhanced Recovery After Surgery protocols, regional anaesthesia, minimally invasive surgery and telehealth. The proposed action plan is structured around the separation of ambulatory patient flows, organization of operational management, economic attractiveness and standardisation of the care pathway. Its implementation requires organizational adjustments and improved community-hospital coordination.Avec un taux de chirurgie ambulatoire de 64 % en 2024, la France demeure en retrait par rapport à l'objectif recommandé de 80 %. Ce mémoire identifie les freins (humains, organisationnels, techniques, financiers) et leviers du développement de la chirurgie ambulatoire. L'étude repose sur l'analyse du cadre réglementaire français, des recommandations institutionnelles, des comparaisons internationales et des observations terrain. Les leviers incluent les protocoles Enhanced Recovery After Surgery, l'anesthésie locorégionale, la chirurgie mini-invasive et la télémédecine. Le plan d'actions proposé comprend la séparation des flux ambulatoires, la structuration de la gestion opérationnelle, l'attractivité économique et la standardisation du parcours patient. Sa mise en œuvre implique des ajustements organisationnels et une meilleure coordination ville-hôpital
Évaluation de points d'intérêt en vision omnidirectionnelle avec une séquence fisheye photoréaliste : Un compte rendu d'expérimentations faites en 2014
No full textWhat is this report:This is a scientific report, contributing with a detailed bibliography, a dataset which we will call now PFSeq for "Photorealistic Fisheye Sequence" and make available at https://doi.org/10.57745/DYIVVU, and comprehensive experiments. This work should be considered as a draft, and has been done during my PhD thesis "Construction of 3D models from fisheye video data-Application to the localisation in urban area" in 2014 [Mor16]. These results have never been published. The aim was to find the best features detector and descriptor for fisheye images, in the context of selfcalibration, with cameras mounted on the top of a car and aiming at the zenith (to proceed then fisheye visual odometry and stereovision in urban scenes). We face a chicken and egg problem, because we can not take advantage of an accurate projection model for an optimal features detection and description, and we rightly need good features to perform the calibration (i.e. to compute the accurate projection model of the camera). What is not this report:It does not contribute with new features algorithm. It does not compare standard features algorithms to algorithms designed for omnidirectional images (unfortunately). It has not been peer-reviewed. Discussions have been translated and enhanced but the experiments have not been run again and the report has not been updated accordingly to the evolution of the state-of-the-art (read this as a 2014 report).</div
Two-Dimensional Digital Electromagnetic Micro-Conveyance Device
No full textInternational audienceThis paper presents a 2D micro-conveyance device based on a 3 × 3 electromagnetic digital actuator array. This device allows the conveyed object to be moved between several discrete positions distributed in the xy-plane through a collaborative actuation of the digital actuators. Each digital actuator includes a mobile permanent magnet placed in a square cavity and can be moved between four discrete positions. An analytical model of the digital actuators was proposed and used to design the conveyance device. Then, a prototype was built using rapid prototyping techniques and was experimentally characterized. The reachable workspace of the conveyance device is 56 mm × 56 mm in the xy-plane, and the proposed architecture enables the workspace to be easily enlarged by adding elementary modules. The distance between two discrete positions is 4 mm, and the positioning repeatability was measured as 5.5 µm. The maximum conveyance velocity and transportable mass were found to be up to 16 mm.s−1 and 15 g, respectively
EsurvFusion: An evidential multimodal survival fusion model based on Epistemic random fuzzy sets
No full textCodes are available online. [Online]. Available: https://github.com/iWeisskohl/EsurvFusion_TFSInternational audienceMultimodal survival analysis aims to combine heterogeneous data sources to improve the prediction quality of survival outcomes. However, this task is particularly challenging due to high heterogeneity and noise across data sources. Additionally, the exact survival time is often censored (partially known) due to incomplete event observation. To address the above challenges, we propose a novel interpretable evidential multimodal survival fusion model, EsurvFusion. This model is designed to combine multimodal data at the decision level using Epistemic Random Fuzzy Sets that jointly handle both data and model uncertainty while incorporating modality-level reliability. Specifically, EsurvFusion first models unimodal data with newly introduced Gaussian random fuzzy numbers, producing possible unimodal survival predictions along with corresponding aleatory and epistemic uncertainty. It then estimates modality-level reliability through a reliability discounting layer to correct the misleading impact of noisy data modalities. Finally, a multimodal evidence fusion layer is introduced to combine the discounted predictions, revealing modality-level influence based on the learned reliability coefficients. Extensive experiments on four multimodal cancer survival datasets demonstrate the effectiveness of our model in handling highly heterogeneous data, establishing a new state-of-the-art performance on several benchmarks
How Well Do LLMs Predict Prerequisite Skills? Zero-Shot Comparison to Expert-Defined Concepts
No full textInternational audiencePrerequisite skills -foundational competencies required before mastering more advanced concepts -are important for supporting effective learning, assessment, and skillgap analysis. Traditionally curated by domain experts, these relationships are costly to maintain and difficult to scale. This paper investigates whether large language models (LLMs) can predict prerequisite skills in a zero-shot setting, using only natural language descriptions and without task-specific finetuning. We introduce ESCO-PrereqSkill, a benchmark dataset constructed from the ESCO taxonomy, comprising 3,196 skills and their expert-defined prerequisite links. Using a standardized prompting strategy, we evaluate 13 state-of-the-art LLMs, including GPT-4, Claude 3, Gemini, LLaMA 4, Qwen2, and DeepSeek, across semantic similarity, BERTScore, and inference latency. Our results show that models such as LLaMA4-Maverick, Claude-3-7-Sonnet, and Qwen2-72B generate predictions that closely align with expert ground truth, demonstrating strong semantic reasoning without supervision. These findings highlight the potential of LLMs to support scalable prerequisite skill modeling for applications in personalized learning, intelligent tutoring, and skill-based recommender systems
The Impact of Thermal and Electrical Pretreatments and Antibrowning Solution on the Chlorogenic and Dicaffeoylquinic Acid Extraction Yield from Endive Roots
No full textInternational audienceForced endive roots (FERs) contain beneficial antioxidant compounds such as chlorogenic acid (5-CQA) and dicaffeoylquinic acids (diCQAs). This study compared the extraction yields of 5-CQA and diCQAs using a biomass pressing method with various pretreatments, including pulsed electric field (PEF) and microwave (MW), against the solid–liquid extraction method (water, 90 °C, 30 min). The results indicated that the MW pretreatment achieved the highest yields, extracting 28 ± 2% of 5-CQA and 13 ± 1% of diCQAs, surpassing the solid–liquid method. Furthermore, the oxidative degradation of CQAs was studied, and it appeared that this reaction was enhanced by PEF pretreatment. An antibrowning solution (ABS) was successfully tested to reduce this oxidation and protect CQAs. An extraction process utilizing MW and PEF pretreatments combined with an ABS solution achieved yields of 65 ± 1% for diCQAs and 80 ± 5% for 5-CQA, significantly outperforming the solid–liquid extraction method
Effects of pulsed electric field and freezing pretreatments on structure and physicochemical properties of different crystalline starches
No full textInternational audienc
Modélisation de la robustesse d'un système informatique par la théorie des catégories
No full textAs part of the study of computer vulnerabilities, we introduce a new deductive system based on sketch theory. We defend two main theses. First, deduction can be interpreted either as a process of forgetting or as its inverse. This leads to a deductive system based on the application of a precomposition functor or its inverse. Second, information is structured data. This perspective allows us to define three preorders to compare information, structure, and data. Our approach prioritizes simplicity. We thus introduce higher-order sketches, which are simply sketches equipped with distinguished tripods (destined to become exponential objects in a model). We demonstrate that any higher-order algebraic structure can be sketched by a higher-order sketch. We characterize strictly deductive sketch morphisms, i.e., those whose associated precomposition functor is invertible and thus allow to establish logical consequences. To achieve this, we generalize to higher-order sketches a process by Andrée and Charles Ehresmann which produces the realization of a sketch, and we introduce a new sketch saturation process. Based on these theoretical foundations, we have developed a concrete programming language : Poule. This language provides an extremely high level of syntactic security at the cost of increased programming complexity. Finally, we explain how computer vulnerabilities can be described using sketches and interpreted as structural defects in a program. We thus propose a formal framework for representing vulnerabilities and design a programming language in which they are avoided through a particularly strict syntax.Dans le cadre de l'étude des vulnérabilités informatiques, nous introduisons un nouveau système déductif fondé sur la théorie des esquisses. Nous défendons deux thèses principales. Premièrement, la déduction peut être interprétée soit comme un processus d'oubli, soit comme son inverse. Cela se traduit par un système déductif basé sur l'application d'un foncteur de précomposition ou de son inverse. Deuxièmement, l'information est une donnée structurée. Cette perspective nous permet de définir trois préordres afin de comparer l'information, la structure et la donnée. Notre approche privilégie la simplicité. Nous introduisons ainsi les esquisses d'ordre supérieur, qui ne sont autres que des esquisses enrichies d'objets exponentiels en devenir. Nous démontrons que toute structure algébrique exprimée en logique d'ordre supérieur peut être esquissée par une esquisse d'ordre supérieur. Nous caractérisons les morphismes d'esquisse strictement déductifs, c'est-à-dire ceux dont le foncteur de précomposition associé est inversible et qui permettent donc de faire des déductions. Pour ce faire, nous généralisons aux esquisses d'ordre supérieur un procédé d'Andrée et Charles Ehresmann permettant de réaliser une esquisse, et nous introduisons un nouveau procédé, appelé saturation des esquisses. À partir de ces travaux théoriques, nous avons créé un langage de programmation concret : Poule. Ce langage offre un niveau de sécurité syntaxique extrêmement élevé, au prix d'une plus grande complexité de programmation. Enfin, nous expliquons comment les vulnérabilités informatiques sont esquissables et peuvent être interprétées comme des défauts de structure d'un programme. Nous proposons ainsi un cadre formel de représentation des vulnérabilités et concevons un langage de programmation dans lequel celles-ci sont évitées grâce à une syntaxe particulièrement stricte