Archives ouvertes de Paris-Saclay
Not a member yet
272730 research outputs found
Sort by
RAVE: Rate-Adaptive Visual Encoding for 3D Gaussian Splatting
No full textInternational audienceRecent advances in neural scene representations have transformed immersive multimedia, with 3D Gaussian Splatting (3DGS) enabling real-time photorealistic rendering. Despite its efficiency, 3DGS suffers from large memory requirements and costly training procedures, motivating efforts toward compression. Existing approaches, however, operate at fixed rates, limiting adaptability to varying bandwidth and device constraints. In this work, we propose a flexible compression scheme for 3DGS that supports interpolation at any rate between predefined bounds. Our method is computationally lightweight, requires no retraining for any rate, and preserves rendering quality across a broad range of operating points. Experiments demonstrate that the approach achieves efficient, high-quality compression while offering dynamic rate control, making it suitable for practical deployment in immersive applications. The code is available at https://github.com/inspiros/RAVE.</div
Validation of Synthetic Method Phase Equilibria Measurements: Re-examination of CO₂ + Methylcyclopentane, CO₂ + 2,2-Dimethylbutane, and Benchmarking with CO₂ + Toluene
No full textInternational audienceRecent publications have reported phase equilibrium data for CO₂ + methylcyclopentane (MCP) and CO 2 + 2,2dimethylbutane using static-analytical sampling methods that are inconsistent with previously published synthetic-method data. To verify the reliability of our earlier measurements, we have repeated vapor-liquid equilibrium (VLE) experiments for both CO₂ + MCP and CO₂ + 2,2-dimethylbutane in the range 20-90 • C using a high-pressure variable-volume PVT cell with visual observation. Our new data agree within 1-2 bar with the values previously reported by our group, confirming the reproducibility of the synthetic method. To further validate our methodology, we also investigated the CO₂ + toluene system, which has been extensively studied in the literature. Our experimental results are in agreement with reference data, thereby confirming the accuracy of the synthetic technique and of the procedures used for mixture preparation and bubble-point detection. These results support the conclusion that the discrepancies between our data and those obtained by static-analytical methods cannot be attributed to errors inherent to the synthetic technique.</div
Modeling the Vibrational Circular Dichroism Spectroscopy of Phenylcyclohexanediol Solvated in Dimethyl Sulfoxide Using Polarizable Molecular Dynamics
No full textInternational audienceModeling vibrational circular dichroism (VCD) spectra in solution remains a complex task due to the intricate interactions involved. Recent advances in computational chemistry, particularly the implementation of polarizable force fields, have greatly improved the description of environmental effects in classical molecular dynamics simulations. In this study, we use the AMOEBA polarizable model to calculate VCD spectra through the expression of the electric and magnetic dipole moments developed recently [Bowles J. et al. ChemPhysChem 2024, 25, e202300982] in combination with the use of induced dipole models. This methodology enables the reliable treatment of flexible molecules and has been applied to the 1-Phenyl-1,2-Cyclohexanediol (PC) molecule in dimethyl sulfoxide (DMSO). The computed infrared and VCD spectra provide detailed insights into molecular conformations and solute-solvent interactions. An in-depth investigation was conducted into the possible effects of system size, concentration, and spectral convergence. Comparison with experimental vibrational frequencies in deuterated solvent further supports the interpretation of spectral features, although residual differences in the spectral shifts and broadenings prove to be challenging for theory, especially in the case of VCD.</div
Pulmonary Hypertension Associated With Left Heart Disease: Challenges, Emerging Strategies, and Future Directions
No full textInternational audiencePulmonary hypertension (PH) associated with left heart disease (PH-LHD) is the most prevalent form of pulmonary hypertension and is associated with substantial morbidity and mortality risk. Diagnosis is challenging due to overlapping phenotypes, imprecise echocardiographic estimates, and reliance on invasive hemodynamics. Pulmonary arterial hypertension (PAH) drugs have shown no benefit and may even be harmful in this population. There are no pharmacologic or interventional therapies approved specifically for PH-LHD to date, other than treatments directed at the underlying LHD. Multiple pharmacologic agents, such as sotatercept, relaxin analogues, and levosimendan, along with device-based strategies like pulmonary artery denervation, transcatheter atrial shunts, and pericardiotomy, are under investigation. Progress in clinical trials has been hindered by patient recruitment, phenotypic heterogeneity, and a lack of standardized functional and hemodynamic outcomes. This review summarizes the current knowledge of PH-LHD, highlights emerging pharmacologic and invasive therapies, discusses barriers to trial design and regulatory approval, and proposes a framework to guide future investigations.</div
Long-term outcomes and temporal trends following liver transplantation for chronic liver disease in intensive care unit.
No full textInternational audienceBackground and Aims: Liver transplantation (LT) remains the definitive treatment for patients with end-stage chronic liver disease (CLD). However, those transplanted while in intensive care unit (ICU) represent a high-risk population. Large-scale data on long-term prognosis in this group are limited. We aimed to assess long-term outcomes in CLD patients undergoing LT from the ICU and to compare outcomes over time.Methods: This retrospective cohort study used the French national transplant registry (CRISTAL). Adults with CLD who underwent LT from the ICU between 2008 and 2018 were included. Organ failures were defined according to EASL CLIF-OF criteria. Five-year survival and associated risk factors were analyzed and compared across four time periods (2008–2010, 2011–2013, 2014–2016, 2017–2018).Results: Among 13,372 LTs performed in France during the study period, 9,686 were for CLD, of whom 1,287 (13.2%) were in the ICU at the time of LT. Alcohol-related liver disease (50%) and viral hepatitis (16.1%) were the leading etiologies. Five-year survival was significantly lower in CLD-ICU patients compared to non-ICU patients (69.2% vs. 79.1%, p<0.0001). Among 1-year survivors, 5-year survival exceeded 83% and was not different to that of CLD patients transplanted outside ICU. Survival by era showed no significant improvement (p=0.28). Age (HR 1.03, p<0.0001), mechanical ventilation (HR 1.57, p=0.0001) and French Donor Risk Score (HR 1.05, 95%CI [1.02-1.09], p<0.001) were independent predictors of mortality.ConclusionPatients transplanted from the ICU have significantly lower long-term survival, primarily due to elevated early post-transplant mortality, with no observed improvement over time. Careful candidate evaluation and donor selection remain critical to improving outcomes in this high-risk population.Impact and Implications: This large national cohort study provides a comprehensive evaluation of long-term outcomes in critically ill cirrhotic patients undergoing liver transplantation while in the ICU. Despite advances in transplant care over the past decade, we observed a persistent survival gap in this high-risk population, primarily driven by increased mortality within the first year after transplantation. Age, the need for mechanical ventilation, and donor-related factors were independently associated with this excess risk. We also found no significant improvement in outcomes over time. These results underscore the continued need for refined candidate selection and donor allocation strategies, taking into account age, pre-transplant clinical stability, and graft quality to optimize post-transplant survival.Trial Registration: ClinicalTrials.gov number NCT0663640
Autocorrelation-based Fiducial Markers for Traceability
No full textInternational audienceClassical approaches to the rectification of a single image of a product, without stereo correspondences, require spatial landmarks. These landmarks, constructed from high-contrast elementary shapes that can be detected with simple algorithms, are highly conspicuous. To rectify complex deformations, one can use chessboard patterns of markers with elements that break quadrilateral symmetry, such as the three eyes of a QR code. However, these marker boards are even more conspicuous than a single marker. In traceability applications, only one site of marking is used, limiting the complexity of the surface on which it can be read, and exposing the mark to deidentification attacks for diversion of the product to a grey market. We introduce a method for constructing stealth and robust fiducial markers that can be displayed across a surface, limiting exposure to marker tampering for product deidentification. These markers, which we refer to as self-rectifying textures, can be used to rectify complex deformations by solving an inverse problem rather than relying on pixel correspondences of conspicuous landmarks. These stealth textures place fiducial markers in the autocorrelation of the image. In this way, crops of the deformed texture can be rectified using only these spatially invariant statistical properties. Affine transformations of an image correspond to linear transformations of the autocorrelation, without phase component. Exploiting this fact, self-rectifying textures enable the local estimation of the differential of a planar deformation by identifying landmarks in the autocorrelation image, such as peaks, whose locations in the fronto-parallel view of the texture are known. The translation component can be recovered independently via phase correlation. A rectifying map, modulo translations, can also be fit directly to local observations of the differential of the deformation, without access to the rectified texture or need for phase correlation. Self-rectifying textures can be used for communication, watermarking, authentication, surface identification, calibration, and geometry processing
Extracting a Toponium Signal at the LHC with Spin and Quantum Information Tools
No full textInternational audienceWe investigate near-threshold top-antitop production at the LHC, focusing on the impact of toponium formation on spin correlations and quantum information properties of the final state. Considering the top-antitop system as a mixed two-qubit state, we reconstruct spin density matrices via quantum tomography and evaluate several observables including some inspired by quantum information. We then compare their sensitivity in discriminating toponium effects from top-antitop production without these effects. Our results demonstrate that combining these variables is expected to significantly enhance sensitivity to toponium effects, bringing new ways to explore these subtle features
Clustering of Exogenous Organic Material Properties to Improve Their Efficient Recycling in European Agriculture
No full textInternational audienceExogenous organic materials (EOMs) are increasingly used as substitutes for mineral fertilizers and as tools to restore soil health within a circular bioeconomy context. However, the great diversity of EOMs in terms of origin, composition, chemical properties, and contaminant concentrations challenges their safe and efficient use in agriculture. The aim of the study was to establish a framework for the clustering of EOMs properties from several EU countries, enabling their categorisation according to their origin, characteristics, and chemical properties and trace elements (TE) profile. For that purpose, a dataset with chemical characteristics and TE concentrations from 118 EOMs was constructed from a database previously published. The EOMs included a wide range of organic residues and waste streams from agricultural, industrial, and urban origins representative of the diversity of European EOMs. Clustering analyses were carried out to distribute EOMs among clusters (i) of chemical properties based on their characteristics (dry mass, C‐to‐N ratio, pH, and concentrations of organic C, N, NNH 4 , P, K, Ca, and Mg), and (ii) of TE profile based on their concentrations of As, Cd, Cr, Cu, Hg, Ni, Pb, and Zn. Five practical scenarios of EOMs applications mixing amending and fertilizing EOMs were considered to estimate input fluxes of C, N, P, K, and TEs over a period of 10 years in contrasted areas (agricultural settings and peri‐urban areas). The present study clustered the EOMs according to their soil improver characteristics that is, from NK fertilizing to liming and organic amending properties. The clustering analysis on TEs classified EOMs according to their TE profiles, with (i) smaller concentration of TEs (i.e., three quarters of EOMs), and (ii) larger concentration of all TEs and especially for Cu and Zn. The various practical scenarios simulating the repeated applications of local EOMs from contrasted areas showed that input fluxes were in line with commercial organic fertilizer inputs and below the goal of 170 kg N per hectare per year, while TE input fluxes respected the French regulation thresholds, even in the scenarios including EOMs containing also greater TE levels
A Complex-Valued Continuous-Variable Quantum Approximation Optimization Algorithm (CCV-QAOA)
No full textContinuous-variable (CV) quantum systems offer a natural framework for continuous optimization through their infinite-dimensional Hilbert spaces. In this paper, we propose the Complex Continuous-Variable Quantum Approximate Optimization Algorithm (CCV-QAOA), a variational framework operating in the complex domain that optimizes over complex decision variables. The method efficiently solves real and complex multivariate optimization problems. To demonstrate its versatility, we apply CCV-QAOA across a broad suite of optimization use cases, including convex quadratic minimization, scaling studies with circuit depth and cutoff dimension, constrained quadratic programs using penalty constructions, and non-convex benchmarks such as the Styblinski-Tang function and complex quartic landscapes
