Portail HAL de l'Université Picardie Jules Verne
Not a member yet
32425 research outputs found
Sort by
Selective Upcycling of Polycarbonate Waste to Cyclohexanol via RuLa Dual‐Atom Catalysis
No full textInternational audienceABSTRACT Cyclohexanol is a key intermediate for nylon manufacture, yet its industrial synthesis relies on partial oxidation of cyclohexane at ∼2 MPa with <5% single‐pass conversion, an energetically wasteful and atom‐inefficient process that demands extensive recycle and generates substantial emissions. Here, we introduce an oxygen‐atom‐efficient synthetic strategy that transforms polycarbonate (PC) waste directly into cyclohexanol through a non‐oxidative catalytic route. The method leverages the intrinsic oxygen functionality of the polymer as a built‐in source of hydroxyl groups, thereby eliminating the conventional oxidation step. A RuLa dual‐atom (RuLa‐DA) catalyst anchored on CoAl oxide enables cooperative hydrogen activation and spillover through moderated Ru–H binding, driving selective aromatic‐ring hydrogenation under mild gas‐phase conditions. Operating at 0.25 MPa and a 4.2 s residence time, the tandem hydropyrolysis–hydrogenation process affords a 69.9% yield and 95.4% selectivity for cyclohexanol, maintaining >95% selectivity for post‐consumer PC over 100 feed cycles. Life‐cycle and techno‐economic analyses indicate the potential environmental and economic advantages, showing a 35% cost reduction and a threefold lower carbon footprint relative to the fossil route. This oxygen‐retentive hydrogenation paradigm establishes a general approach for valorizing oxygen‐rich substrates and suggests a conceptually viable pathway toward atom‐economical synthesis and circular chemical manufacturing
Management of severe asymptomatic aortic stenosis: An updated overview
No full textInternational audienceAortic stenosis is the most frequent valvular heart disease in industrialized countries, affecting about 10% of individuals aged > 75 years, with 2-4% presenting severe disease; nearly half of them are asymptomatic. No medical therapy halts aortic stenosis progression, and aortic valve replacement remains the only curative option. Managing asymptomatic patients is challenging because of a persistent risk of sudden death, and the potential for irreversible myocardial damage if intervention is delayed, but also because aortic valve replacement carries procedural risks and long-term prosthesis-related complications. Functional assessment, especially exercise testing, is essential as it can identify a substantial proportion of "falsely asymptomatic" patients who will benefit from timely aortic valve replacement. Accurate assessment of aortic stenosis severity is crucial, with echocardiography as the cornerstone, although additional imaging modalities may be required in selected cases. Modern risk stratification integrates markers of subclinical myocardial dysfunction, haemodynamic compromise and procedural risk. Although observational studies and randomized trials suggest that early aortic valve replacement may reduce adverse events in selected low-risk patients, results are heterogeneous and methodological concerns remain. Meta-analyses have shown reductions in unplanned hospitalizations but no clear survival benefit with early intervention. Current guidelines recommend aortic valve replacement for classical indications, such as symptom onset or reduced ejection fraction, and suggest early aortic valve replacement only in carefully selected asymptomatic patients with low procedural risk. Shared decision-making within a multidisciplinary heart valve team remains essential, taking into account patient preferences, co-morbidities, life expectancy and lifestyle. Close follow-up with regular echocardiography and functional testing is critical to optimize timing and outcomes. (c) 2025 The Authors. Published by Elsevier Masson SAS. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)
Fed-SHAP-IDS: Federated SHAP-Based Intrusion Detection System for IoMT
No full textInternational audienceThe use of the Internet of Medical Things (IoMT) has significantly increased in recent years, introducing new vulnerabilities and security challenges. Numerous attacks now target the healthcare ecosystem, exploiting weaknesses in connected medical devices. Traditional intrusion detection methods are increasingly insufficient to cope with the growing number of attacks and the emergence of zero-day threats. Artificial Intelligence (AI) has demonstrated strong potential in this context; however, centralized AI-based detection solutions expose sensitive medical data, create a single point of failure, and struggle with non-IID (non-independent and identically distributed) data across geographically dispersed devices. To overcome these limitations, we propose Fed-SHAP-IDS (Federated SHAP-based Intrusion Detection System), a novel federated learning framework specifically designed for IoMT environments. This system preserves data confidentiality by training models locally without sharing raw data, while effectively handling non-IID data distribution across heterogeneous clients. Furthermore, Fed-SHAP-IDS integrates the SHAP (Shapley Additive Explanations) method to capture local feature importance, enabling interpretable, explainable, and interoperable insights for each client. By combining federated learning with SHAP-driven feature aggregation, our approach enhances both privacy and transparency, offering a robust and trustworthy solution for IoMT intrusion detection
Prognosis of critically ill patients with severe acute kidney injury and high circulating dipeptidyl peptidase 3: a post hoc analysis of the AKIKI 2 trial
No full textInternational audienceNo abstract availabl
« Une fleur ne fait pas le printemps. Pour une vie heureuse, il faut beaucoup d'enfants.» La construction de la « crise démographique » comme problème public en Iran
No full textInternational audienceComment les politiques natalistes dans l’Iran contemporain génèrent-elles des pratiques de contrôle différencié des naissances selon les appartenances de classe et ethnoraciale ? De quelle manière les dispositifs de santé reproductive contribuent-ils à perpétuer et à renforcer les hiérarchies sociales, de genre et raciales existantes ?Focalisée sur ces questions, ma communication examine la construction de la « crise démographique » comme problème public en Iran. Elle analyse la genèse et la mise en œuvre des politiques reproductives, en particulier leur application différenciée selon l’appartenance sociale et ethnoraciale des femmes. J’illustre notamment les effets discriminatoires systémiques de la loi 315, dite « Plan pour la jeunesse et le soutien à la famille » de 2014, révélant des mécanismes complexes de contrôle biopolitique.Mon analyse s’appuie sur une approche intersectionnelle comme socle de réflexion. Celle-ci permet d’articuler les rapports de genre, de classe et d’ethnie afin de comprendre les expériences reproductives différenciées. Cette grille critique déconstruit les visions homogénéisantes des femmes iraniennes et met en lumière la manière dont les politiques de santé reproductive marginalisent spécifiquement les femmes issues des minorités ethnoraciales ou vivant dans des régions périphériques.L'apport principal de cette communication réside dans la mise en évidence d’une alliance stratégique entre l’État et la profession médicale, qui structure l’ensemble des politiques de population mises en œuvre en Iran depuis la révolution islamique. Cette collaboration institutionnelle perdure malgré les alternances politiques.Derrière les réorientations — tantôt natalistes, tantôt restrictives — se maintient une remarquable continuité des modalités opérationnelles. Les politiques de limitation des naissances mises en place après la révolution illustrent cette coopération. Comme l’a expliqué un gynécologue rencontré sur mon terrain de recherche, des unités sanitaires mobiles, composées de professionnel·le·s médicaux·les, déployaient des campagnes ciblées dans les zones rurales. Ces agents mobilisaient leur autorité pour légitimer les recommandations contraceptives, mettant la confiance accordée au corps médical au service des objectifs étatiques.Le tournant nataliste amorcé dans les années 2000 traduit une institutionnalisation accrue de cette alliance. Le corps médical ne se contente plus d’exécuter les directives : il participe à leur élaboration. Pour renforcer sa position, il a contribué à la construction du discours sur une prétendue « crise démographique » — un concept, comme l’ont montré Mona Claro (2018) et Elisabeth Krause (2018), plus performatif que descriptif, visant à légitimer un pouvoir d’expertise. Cette dynamique s’accompagne d’une médicalisation des enjeux démographiques : les problèmes sociaux sont reformulés comme des questions de santé publique, appuyées par un usage stratégique des données statistiques
Data-driven machine learning modelling for the manufacturing of the fuel electrode support in solid oxide cells
No full textInternational audienceThe industry-relevant fabrication of supports in fuel-electrode supported Solid Oxide Cells (SOCs) by tape casting typically involves a multi-stage process, demanding precise control over tape thickness and density. However, conventional SOC manufacturing processes are resource-intensive and often rely on industry/R&D unpublished knowledge and trial-and-error practices to achieve the target properties of the resulting tape. Hence, machine learning (ML) was employed for predicting the thickness and density across three distinct stages of the fabrication process: tape casting, sintering, and NiO-reduction process. Our developed ML models (e.g., Extra Trees and Ridge Regressions) demonstrate exceptional accuracy (R-2 > 0.9) for each specific prediction task. Concurrently, experimental data analysis was conducted to elucidate the impact of the manufacturing parameters on the tape properties. Our data-driven ML approach offers a pathway towards achieving precise tape property control and advancing more efficient SOC support manufacturing
Uma poética da melancolia: a embriaguez de sair de si em Thomas l’Obscur
No full textInternational audienceO presente ensaio explora Thomas l’Obscur (1950), de Maurice Blanchot, como um espaço de dissolução do ser e de experiência do impossível, articulando-o a uma poética da melancolia e da desaparição. Thomas l’Obscur é uma obra que nos coloca diante da incapacidade de acessar a totalidade do ser, da impossibilidade de se encontrar em uma relação autêntica com o outro, e da eterna busca por um significado que não se realiza. A obra reflete uma experiência do impensável e do irrepresentável, uma escrita que se faz a partir da do limite, da fronteira entre o ser e o não-ser, o visível e o invisível. Desta forma, o objetivo é analisar como a obra encena a fragilidade do sujeito, o esfacelamento da identidade e a recusa da literatura como representação, transformando-a em um campo de forças onde o ser se dispersa no vazio. Para tanto, este ensaio busca fundamentar-se a partir da concepção blanchotiana de literatura como movimento para o neutro e para o silêncio, da problematização do ser e do nada e de sua compreensão da noção de experiência-limite baitalliana, além de utilizarmos como suporte demais leituras que conversem com o tema
Stabilisation multigrille des intégrateurs temporels pour les EDP paraboliques non linéaires
No full textInternational audienceWe propose to enhance the stability of IMEX schemes applied to the numerical solution of nonlinear parabolic equations by using proper smoothing procedures on band of frequency components of the solution. The separation of those different components is realized by considering a multigrid (MG) approach. The time scheme on the coarset grid is implicit and energy diminishing, while the time marching is done by a weighted damped IMEX scheme on the finer grids. This approach allows to concentrate the main computational effort on a small set of components (the coarset grid), and also to tune finely the damping parameters in such a way to have a limited consistency perturbation while enhancing the stability. We establish stability results and give numerical illustrations on pattern dynamics, Swift-Hohenberg and then Crystal Phase Fields equations, in both 2D and 3D cases. The results we obtain with MG are comparable with those given by a one grid simulation.</div
Present-day and future global maps of non-native species richness across 17,343 islands worldwide
No full textIsland biodiversity is increasingly threatened by non-native species (NNS). Yet, no global map of island NNS richness has been produced to support global threat assessments. Using a hurdle random forest model informed by 11 biogeographic and anthropogenic predictors, we predicted NNS richness for 17,343 islands for the present day and the near future (2050). Biogeographic factors were the strongest predictors of NNS establishment, while the combined influence of biogeographic and anthropogenic variables best explained the accumulation of NNS across islands once established. We found NNS richness to be greater for larger, more elevated, and more remote islands, with high human population densities and low natural habitat proportions. Models predicted 54% of islands invaded in 2020, and an increase in NNS richness for 16% of these islands by 2050. These predictions help identifying islands prone to biological invasions, both today and in the future, and thus to derive conservation prioritizations