HAL: Hyper Article en Ligne
Not a member yet
3159010 research outputs found
Sort by
Global and local implications of computational artefacts
No full textInternational audienceThis article investigates the undone science of software scalability. Software engineering literature focuses on the multiple ways that software can scale up, but too rarely on the possibilities, or necessities, of making it scale down. Drawing from geographical studies, we hypothesize that scales are constructed, rather than given, and argue that the design and use of specific computational artefacts-programming languages, data structures and protocols-are active participants in such material construction. Through a series of case studies, we discuss how computational artefacts can realize computing at both local and global scales
Accurate Dating of Ancient Manuscripts with SIFT-CNN
No full textInternational audienceHistorical manuscripts dating remains a significant challenge in the fields of document analysis and classification, and digital preservation. This study aims at investigating a hybrid machine learning model for ancient manuscripts dating. This later integrates the Scale-Invariant Feature Transform (SIFT) with advanced Convolutional Neural Networks (CNNs). To allow the encapsulation of manuscripts intricate characteristics, the model leverages SIFT to extract and encode local features that are afterwards organized into a comprehensive visual codebook. A proposed custom-designed CNN architecture exploits these local features to decipher complex visual patterns indicative of specific historical periods. The deep layers of the network learn hierarchical representations that reflect the evolution of script styles, material composition, and textual formats, enabling the system to detect and interpret chronological signatures within the manuscript imagery. The proposed hybrid model is extensively trained and its performances are assessed on the KERTAS dataset to allow for identifying and exploiting subtle visual markers associated with temporal data. The early results demonstrate the model’s substantial potential to streamline the manuscript dating process, offering historians and archivists a robust tool for more efficient historical document analysis including archival materials cataloging
Collective Analysis of Work to Increase the Amount of Time Spent with Patients. A Case Study Looking at the Value of Activities Using the Lean Approach
No full textInternational audienceFor several years now, healthcare professionals have expressed their frustration at the increasing number of administrative and logistical tasks that take them away from patients’ bedsides, where their work is most meaningful. In March 2022, the French Senate published a report on the state of hospitals and the healthcare system in France. Faced with a “never-ending crisis” (Speech by Emmanuel Macron, presentation of the National Health Plan, 6 January 2023.) and a feeling of loss of purpose widely expressed by healthcare professionals, “we need to put patient care back at the heart of the hospital professions, which are too far removed from it because they do not have enough time to devote to it” (Commission of Inquiry Report No. 587 (2021–2022), Volume I, published on 29 March 2022.). Lean, a method of reorganising processes that focuses on customer-perceived value, would appear to be a promising solution to the situation described above. Based on the following research question: “How can collective analysis of work during lean healthcare implementation contribute to increase the amount of time spent with patients?”, we will explore ways of analysing the activities carried out by healthcare professionals in order to increase the amount of time devoted to care. Our research is based in the geriatric ward of a hospital group. For a number of years, this ward has faced major organisational difficulties that have led to occupational burnout. The results show the potential for refocusing healthcare activities and limiting non-value added tasks, based on an analysis of the tasks performed by professionals in the field
Spatial resolution studies with the BabyIAXO Micromegas prototype
No full textInternational audienceThe spatial resolution of the Micromegas prototype developed for the BabyIAXO experiment was evaluated using a low-energy X-ray beam at the SOLEIL synchrotron facility. BabyIAXO, currently under construction, aims to search for hypothetical solar axions. A key component of the experiment is a low-background X-ray detector with high efficiency in the 1-10 keV energy range and stringent background rejection capabilities. Achieving a spatial resolution on the order of, or better than, 1 mm is critical for accurately reconstructing signal shapes and positions, and for effectively discriminating between signal and background events. Therefore, a precise characterization of the detector's spatial resolution is essential to validate its suitability for the experiment. This study involved scanning the IAXO-D1 Micromegas detector under various beam energies, positions, and drift field configurations to evaluate their influence on spatial resolution. A resolution of approximately 100 m at 6 keV was achieved, confirming the strong potential of this technology for application in the final BabyIAXO setup
I/O patterns modeling of HPC applications with call stacks for predictive prefetch
No full textInternational audienceModern high-performance computing (HPC) storage systems use heterogeneous storage technologies organized in tiers to find a compromise between capacity, performance, and cost. In these systems, prefetching is a common technique used to move the right data at the right moment from a slow to a fast tier to improve overall performance while using the costly high-performance tier only when needed. Effective prefetching requires precise knowledge of the application I/O patterns. This knowledge can be extracted through the source code, I/O tracing tools or I/O functions call stacks. State-of-the-art solutions based on the latter approach mainly focus on applications with regular I/O profiles to avoid scalability issues due to the grammar-based techniques used. In this paper, we present an approach based on I/O call stacks that models POSIX and STDIO I/O patterns for both regular and irregular applications, thanks to the use of directed graphs. We present different models usable for prefetching. Our models were used to predict the next I/O call stack on five real HPC applications with a prediction accuracy of up to 98%. Compared to the state-of-the-art Omnisc'IO, they incurred up to 120x lower model overhead (334 ns vs. 45 μs on LAMMPS) and had a model size 10x to 15x smaller (463 B vs. 7 kB on LQCD)
A Method for Constructing a Digital Twin Monitoring System for Bridge Cranes Based on Unity3D
No full textInternational audienceWith the rapid advancement of intelligent manufacturing, ensuring the safe operation and real-time monitoring of bridge cranes has become increasingly critical. Traditional monitoring methods are often hampered by data silos, limited real-time capabilities, and weak visualization, failing to meet the demands of modern industrial equipment management. To address these challenges, this study developed a digital twin-based monitoring system for bridge cranes using the Unity3D engine. The system integrates 3D modeling, physical simulation, and cyber-physical data interaction to enable real-time visualization and historical reconstruction of operational status. A multi-layer architecture—comprising data acquisition, processing, virtual model construction, and application display—was designed to integrate multi-source sensor data and support real-time mapping of key components such as the trolley, bridge travel mechanism, and hoisting device. The system was validated experimentally using a 20-ton double-girder bridge crane in an industrial indoor environment, demonstrating high-precision synchronization with the physical entity, along with reliable monitoring, historical replay, and anomaly recognition. This approach provides practical value for industry users by enhancing operational safety, facilitating predictive maintenance, and improving intelligent management of special industrial equipment
Groundwater Vulnerability Assessment around Open Dumpsites in a Nigerian Basement Complex Terrain Using a Modified DRASTIC-L Model
No full textInternational audienceThe risk of aquifer contamination is increasing in areas underlain by Basement Complex terrain due to continuous waste disposal. The primary aim of this study is to evaluate the intrinsic and specific vulnerability of groundwater resources to contamination from leachate originating from open dumpsites within a Nigerian Basement Complex terrain by applying a Modified DRASTIC-L model. This study develops a composite aquifer vulnerability map for selected dumpsites using a modified version of the DRASTIC model, referred to as DRASTIC-L. The study area is situated in Itaogbolu, within the Akure North Local Government Area of Ondo State, Nigeria. Unlike the traditional DRASTIC framework, which evaluates seven hydrogeological parameters; depth to groundwater, net recharge, aquifer media, soil media, impact of the vadose zone, and hydraulic conductivity. The DRASTIC-L model incorporates land use as an additional factor. Vulnerability mapping was carried out in ArcGIS using inputs from Vertical Electrical Soundings (VES), remote sensing, and hydrogeological surveys. The VES campaign comprised 56 depth soundings with electrode separations up to 65 m, while remotely sensed parameters were derived from LANDSAT 7 ETM+ imagery and a Digital Elevation Model (DEM). The DRASTIC-L Vulnerability Index (DVI) was calculated by assigning ratings (1–10) and weights (1–5) to each parameter. The study classified the area into five vulnerability zones: very high, high, medium, low, and very low. Findings indicate that several dumpsites are located in high-vulnerability zones, posing significant risks to groundwater quality. Model validation using water sample analysis confirmed an accuracy of 76%, highlighting the importance of improved waste management practices for groundwater protection. Given these findings, the implementation of proper waste and water management policies in the study area is essential
Edible Mushrooms as Functional Foods: Phytochemical Composition, Biological Activities, and Health Applications- A Comprehensive Review
No full textInternational audienceEdible mushrooms are increasingly recognized as functional foods due to their favorable nutritional profile and rich diversity of bioactive compounds. Beyond their traditional culinary value, mushrooms synthesize structurally diverse secondary metabolites through fungal-specific metabolic pathways, contributing to a wide range of health-promoting effects. This review summarizes the nutritional composition of commonly consumed edible mushrooms and critically examines phenolic compounds as key contributors to their biological activity.The macronutrient and micronutrient composition of mushrooms, including carbohydrates, dietary fiber, proteins, lipids, vitamins, and minerals, is discussed alongside the influence of cultivation practices, environmental conditions, and processing methods on bioactive content and bioavailability. Emphasis is placed on the classification and occurrence of mushroom phenolics, including phenolic acids, flavonoid-like compounds, tannins, and related derivatives, highlighting species- and substrate-dependent variability. Current evidence regarding the debated origin and biosynthesis of flavonoids in mushrooms is also reviewed.The biological activities of mushroom phenolics—such as antioxidant, anti-inflammatory, anticancer, and antihyperglycemic effects—are summarized with reference to proposed molecular mechanisms. Interactions between phenolics and other mushroom bioactives, including polysaccharides, tocopherols, terpenoids, and sterols, are considered in the context of functional food and therapeutic applications. Despite promising in vitro findings, limitations related to bioavailability, analytical standardization, and insufficient in vivo and clinical validation remain. This review highlights key research gaps and future directions necessary for the evidence-based development of mushroom-derived functional foods and nutraceuticals
Exploring Sustainability Challenges in Tanzania’s Fast Ferries Transportation System
No full textInternational audienceThe study explored sustainability challenges in Tanzania’s fast ferries transportation system. The study was guided by three predicting variables tested on the sustainability of fast ferries transportation as the dependent variable namely competition, service quality provision and perceived cost. The study was performed using explanatory study design whereas causal relationship testing was conducted to generate new knowledge on the inquired subject. Data were collected from the employees in Azam Marine Company Limited since it is the only company that has survived well for years with prospects of long sustainability useful for information gathering process in the context. The data were assembled and grouped altogether such that were computed in SPSS version 23.0 for the generation of relevant statistical measurements to present the results of the study. Multiple regression analysis was used to show the existing relationship between study variables in filling the inquiry gap on factors affecting sustainability of fast ferries transportation in Tanzania. Study results revealed that among three predicting variables only two namely competition and perceived cost have been generated positive with significant effect statistically on sustainability of fast ferries transportation with p<0.05. However, service quality provision has been generated positive with insignificant effect on the sustainability of fast ferries transportation with p>0.05. This implied that sustainability in fast ferries transportation in Tanzania is affected with competition and perceived cost. It is therefore recommended that the companies in the sector should be rational in cost assessment since it has implications towards the prosperity of the business in terms of returns and sustainability
Safety and Tolerability of Twice-Yearly Depemokimab in Patients with Asthma and Chronic Rhinosinusitis with Nasal Polyps: Pooled Results from SWIFT-1/-2 and Anchor -1/-2
No full textInternational audienceIntroduction: Depemokimab, the first ultralong- acting respiratory biologic, is under investigation for diseases with underlying type 2 (T2) inflammation. Subcutaneous depemokimab 100 mg was efficacious in patients with T2 asthma characterized by blood eosinophil count and in chronic rhinosinusitis with nasal polyps (CRSwNP) when administered twiceyearly in four Phase III randomized, doubleblind, placebo-controlled studies (SWIFT-1/-2 and ANCHOR-1/-2, respectively). This pooled analysis examined the safety and tolerability of subcutaneous depemokimab 100 mg versus placebo in patients with T2 asthma and in CRSwNP. Methods: Adverse event (AE) frequency, severity, time-to-onset, duration, and relative risk (RR) were evaluated for depemokimab 100 mg and placebo administered subcutaneously every 26 weeks for 52 weeks using pooled SWIFT-1/-2 and ANCHOR-1/-2 data. The impact of immunogenicity on safety was also evaluated.Results: Overall, 1290 patients from the four studies were included in the safety analysis population (773 [60%] received depemokimab; 517 [40%] received placebo). On-treatment AEs were reported by 73% and 78% of patients in the depemokimab and placebo groups, respectively; the majority were mild or moderate in intensity and transient, with similar durations between treatment groups. Serious AEs (SAEs; 5% and 10%, with depemokimab and placebo, respectively), and discontinuations due to AEs (< 1% and 1%, respectively) were infrequent. No fatal AEs or treatment-related SAEs (per investigator assessment) were reported. RRs (vs. placebo) were similar for all common on-treatment AEs except asthma and back pain, which occurred less frequently in the depemokimab group compared with the placebo group. Antidrug antibodies and neutralizing antibodies occurred infrequently, and no association between antidrug antibody status and depemokimab efficacy was identified.Conclusions: In these studies, twice-yearly depemokimab 100 mg was generally well tolerated by patients with T2 asthma or CRSwNP over the 52-week treatment period, supporting the safety of the first ultra-long acting biologic for these diseases.Clinical trial identifier(s): NCT04719832/ NCT04718103/NCT05274750/NCT05281523