Portail HAL UNIV-RENNES
Not a member yet
180610 research outputs found
Sort by
Comparaison des Impacts Environnementaux des Réseaux de Distribution AC, DC et Mixte
No full textInternational audienceLa croissance des besoins en électricité a entraîné une augmentation du nombre d'appareils à courant continu (DC), tels que les panneaux photovoltaïques et les chargeurs de véhicules électriques. Cela ravive la recherche autour d'un réseau de distribution DC, ou du moins dans un premier temps mixte AC/DC. La comparaison entre ces alternatives doit prendre en compte les impacts environnementaux associés aux dispositifs électroniques de puissance spécifiques. Cet article propose une comparaison des impacts d'installation et d'exploitation de trois choix techniques pour les réseaux électriques à basse tension : DC, AC et mixte AC/DC. Une étude de cas est choisie sur la base de sources d'énergie distribuées mixtes typiques de consommations. Des simulations sont effectuées sur une période de 20 ans pour chacune des trois architectures. Cette étude montre qu'une topologie mixte peut être une alternative prometteuse aux réseaux traditionnels à courant alternatif, car elle réduit l'impact des convertisseurs, en particulier pour les courtes longueurs de câble. Cependant, l'architecture optimale pour un système donné peut varier en fonction de facteurs tels que la puissance nominale, la longueur du câble et le profil de consommation. Pour explorer cette variabilité, une analyse de sensibilité détaillée est réalisée.Mots-clés -réseau DC, réseau AC, réseau mixte AC/DC, impacts environnementaux, éco-conception, éco-dimensionnement.</p
Carrié (Fabien), Traïni (Christophe), S’engager pour les animaux , Paris, Presses universitaires de France, coll. « La vie des idées », 2019, 120 p.
No full textInternational audienc
153P Evolution of second-line practices in the era of immunotherapy: Real-life data from the French prospective CHIEF cohort
No full textInternational audienceBackgroundAdvanced hepatocellular carcinoma (HCC) is a global health challenge, and while immunotherapy-based combination are now standards of care as first-line treatment, there is a paucity of data regarding second-line treatment following immunotherapy. We aimed to study the access to second-line treatment, comparing patients initially treated with Sorafenib (Sor) to those with atezolizumab-bevacizumab (AB).MethodsThis study included patients from the real-world prospective CHIEF cohort and aimed to get insight into Systemic treatment sequences in patients with advanced HCC (STRETCH). Median overall survival (mOS) and median progression free survival (mPFS) were defined since the beginning of the 2d line setting.ResultsBetween September 2019 and September 2024, 1103 patients received either AB (n=899) or Sor (n=204) as first-line treatment. Most patients were Child-Pugh A (77.1% AB vs. 70.3% Sor; p=0.06); BCLC-C (66.3% AB vs. 84.3% Sor; p<0.001) and ALBI grade 2 (61.2% AB vs. 64.9% Sor; p=0.021). mOS and mPFS were 22.3 [18.5–27.4] and 5.6 [5.2–6.4] months for AB, compared to 9.4 [7.3–12.7] months (p<0.0001) and 3.5 [3.1–4.0] months (p=0.11) for Sor. Among 456 patients progressing on AB, 42.1% received second-line treatment versus 60.0% of 130 patients progressing on Sor (p<0.0003). In second-line, after AB, patients receiving tyrosine kinase inhibitors (TKI) had an mOS of 13.0 [9.8–15.5] months (n=136, 70.8%), while those on immunotherapy (n=24) or combinations (n=24) had non-reached mOS (p=0.00088). The mOS for second-line TKI was similar regardless of first-line treatment (8.6 months after Sor; p=0.082). No significant differences were found between Sor (n=78), lenvatinib (n=35), regorafenib, or cabozantinib (n=23) (p=0.83), though lenvatinib showed a trend for improved mPFS.ConclusionsThis unique prospective cohort provides real-world data on second-line treatment practices. Access to second-line treatment was lower after AB than Sor, but survivals under second-line TKI were similar. While second-line immunotherapy following immunotherapy shows promising results, these are likely influenced by patient selection for rechallenge
SocialForge: simulating the social internet to provide realistic training against influence operations
No full textInternational audienceSocial media platforms have enabled large-scale influence campaigns, impacting democratic processes. To fight against these threats, continuous training is needed. A typical training session is based on a fictive scenario describing key elements which are instantiated into a dedicated platform. Such a platform simulates social networks, which host a huge amount of content aligned with the training scenario. However, directly using Large Language Models to create appropriate content results in low content diversity due to coarse-grained and high-level scenario constraints, which compromises the trainees' immersion. We address this issue with SocialForge, a system designed to enhance the diversity and realism of the generated content while ensuring its adherence to the original scenario. Specifically, SocialForge refines and augments the initial scenario constraints by generating detailed subnarratives, personas, and events. We assess diversity, realism, and adherence to the scenario through custom evaluation protocol. We also propose an automatic method to detect erroneous constraint generation, ensuring optimal alignment of the content with the scenario. SocialForge has been used in real trainings and in several showcases, with great end-user satisfaction. We release an open-source dataset 1 generated with SocialForge for the research community
MVAE : Motion-conditioned Variational Auto-Encoder for tailoring character animations
No full textInternational audienceThe design of character animations with enough diversity is a time-consuming task in many productions such as video games or animated films, and drives the need for more simple and effective authoring systems. This paper introduces a novel approach, a motion-conditioned variational autoencoder (VAE) with Virtual reality as a motion capture device. Our model generates diverse humanoid character animations only based on a gesture captured from two Virtual reality controllers, allowing for precise control of motion characteristics such as rhythm, speed and amplitude, and providing variability through noise sampling. From a dataset comprising paired controller-character motions, we design and train our VAE to (i) identify global motion characteristics from the movement, in order to discern the type of animation desired by the user, and (ii) identify local motion characteristics including rhythm, velocity, and amplitude to adapt the animation to these characteristics. Unlike many text-tomotion approaches, our method faces the challenge of interpreting high-dimensional, non-discrete user inputs. Our model maps these inputs into the higher-dimensional space of character animation while leveraging motion characteristics (such as height, speed, walking step frequency, and amplitude) to fine-tune the generated motion. We demonstrate the relevance of the approach on a number of examples and illustrate how changes in rhythm and amplitude of the input motions are transferred to coherent changes in the animated character, while offering a diversity of results using different noise samples
Relugolix in prostate cancer therapy: Clinical evidence and practical considerations
No full textInternational audienceIntroduction: Androgen deprivation therapy (ADT) remains a cornerstone of treatment for both localized and metastatic prostate cancer (PC). Relugolix, an oral gonadotrophin-releasing hormone antagonist, provides a new option for achieving rapid testosterone suppression using an oral formulation.Materials and methods: A comprehensive literature search was conducted in PubMed by combining the search terms \"relugolix\", \"TAK-385\", \"MVT-601\", \"prostate cancer\", and \"prostatic neoplasms\" and focusing on prospective and retrospective studies published in English.Results: The HERO pivotal phase III trial demonstrated sustained testosterone suppression in 96.7% of patients with PC treated with relugolix versus 88.8% with leuprolide through to 48weeks (P<0.001). Relugolix achieved faster testosterone suppression and recovery post-treatment and a 54% lower risk of major adverse cardiovascular events compared with leuprolide. Data from HERO and phase II studies also support its use in combination with radiotherapy or other systemic therapies. Real-world studies performed to date have confirmed the effectiveness of relugolix, with more than 98% patients achieving castrate testosterone levels. Adherence to relugolix was generally high, and its safety profile aligned with clinical trial data. Practical considerations include, among others, treatment combination and drug-drug interactions, patient choice, and oncological outcomes.Conclusions: Clinical trials and real-world evidence support relugolix as a convenient and effective ADT option for PC. It offers rapid and sustained testosterone suppression, potential cardiovascular benefits, and an alternative to injectable therapies. Long-term adherence, the use of combination treatments and related drug-drug interactions require further investigation
Adverse reactions of liquorice consumed in the diet: a 10-year retrospective study of poison centres in France
No full textInternational audienceIntroduction: We aimed to describe the symptoms, patient demographics, and trends over time of adverse effects related to liquorice consumed in the diet reported to French poison centres.Methods: We performed a retrospective study of data from French poison centres of cases of adverse effects of liquorice consumed in the diet, with a high causality score, between 2012 and 2021 (10 years).Results: Sixty-four cases were included. The annual number of cases ranged from three to nine, with no significant variation over the study period. Liquorice-induced reactions were very rare (0.008% of all cases with symptoms reported to French poison centres). The products consumed were non-alcoholic beverages (non-alcoholic pastis, liquorice-based Antésite®, and liquorice syrup), alcoholic beverages of the pastis type (10.9%), confectionery containing liquorice (12.5%), confectionery made with liquorice extract only (9.4%), herbal teas (12.5%) and food supplements (4.7%). Consumption was commonly chronic (67.2%) and non-compliant (70.3%). Chronic users presented with symptoms suggestive of pseudohyperaldosteronism, the severity of which seemed to correlate with the amount of glycyrrhizin ingested. Severity was high in 43.8% of cases. When the outcome was known (56.3%), it was favourable in almost all cases (94.4%), often after inpatient care, particularly in an intensive care unit. One patient had sequelae due to a stroke, and one fatality was reported. Severe cases were observed with all types of products, except liquorice syrup and food supplements, and more frequently with beverages (pastis with or without alcohol, and Antésite®).Discussion: Due to significant variability in response to glycyrrhizin, some patients presented signs and symptoms suggestive of pseudohyperaldosteronism such as hypokalaemia, salt and water retention, and hypertension despite consuming the product as directed.Conclusions: Liquorice-induced effects were rarely reported to French poison centres, but their severity was high. Most patients were adults with chronic and non-compliant consumption, especially of soft drinks, with a clinical presentation suggestive of pseudohyperaldosteronism
ADRC-based cascade controller applied to Stacked Interleaved Buck Converter Powering a PEM Electrolyser: Comparison with backstepping method and PI compensator
No full textInternational audienc
Comparative Evaluation of Robust Control Strategies for Grid-Tied Solar Generator Under External Disturbances
No full textInternational audienc
DiLiPop — Code affilié à Bouvrais et al. EMBO Reports (2021)
No full textThis code depends on 3 modules also available at zenodo with DOIsdoi.org/10.5281/zenodo.4552493 whose content is to be placed in Tracking_Matlab/Tracking_Matlab/rdir/denoisingdoi.org/10.5281/zenodo.4552491 whose content is to be placed in Tracking_Matlab/Tracking_Matlab/rdir/BayesianInferencedoi.org/10.5281/zenodo.4552487 whose content is to be placed in Tracking_Matlab/Tracking_Matlab/rdir/treatment_at_the_cortexA brief documentation is offered in BayesianInference/How to use DiLiPop code.pdf (thus in the doi.org/10.5281/zenodo.4552491 module