A5022637848- Bayesian Methods and Mixture Models
- Statistical Methods and Inference
- Medical Imaging Techniques and Applications
- Radiation Detection and Scintillator Technologies
- Nuclear Physics and Applications
- Nuclear reactor physics and engineering
- Advanced Radiotherapy Techniques
- Fault Detection and Control Systems
- Radioactivity and Radon Measurements
- Graphite, nuclear technology, radiation studies
- Nuclear and radioactivity studies
- Geochemistry and Geologic Mapping
- Distributed Sensor Networks and Detection Algorithms
- Gaussian Processes and Bayesian Inference
- Particle Detector Development and Performance
- Radiomics and Machine Learning in Medical Imaging
- Statistical Methods and Bayesian Inference
- Markov Chains and Monte Carlo Methods
- Statistical and numerical algorithms
- Radiation Dose and Imaging
- Radiation Therapy and Dosimetry
- Advanced Statistical Process Monitoring
- Radioactive contamination and transfer
- Spectroscopy and Chemometric Analyses
- Cold Atom Physics and Bose-Einstein Condensates
Commissariat à l'Énergie Atomique et aux Énergies Alternatives
2015-2024
CEA LIST
2014-2024
Université Paris-Saclay
2021-2024
Moscow Power Engineering Institute
2023
Maison de la Simulation
2013-2022
CEA Paris-Saclay
2006-2022
Integra (United States)
2009-2014
Organisation de Coopération et de Développement Economiques
2006
CEA DAM Île-de-France
2006
CEA Paris-Saclay - Etablissement de Saclay
2005
This paper proposes an extension of standard mixture stochastic models, by replacing the constant weights with functional defined using a classifier. Classifier Weighted Mixtures enable straightforward density evaluation, explicit sampling, and enhanced expressivity in variational estimation problems, without increasing number components nor complexity components.
This article presents the synthesis and blend of bismuth complexes in polystyrene based plastic scintillators. A specific design has enabled fabrication a scintillator loaded with up to 17 wt% bismuth. Tri-carboxylate triaryl compounds were used explore understand influence loading on two main criteria scintillation: light yield detection efficiency γ-rays. For gamma radiation an energy <200 keV, scintillators demonstrate ability produce photoelectric peak (total absorption peak) pulse...
The detection and estimation of filtered point processes using noisy data is an essential requirement in many seismic, ultrasonic, nuclear applications. We address this joint detection/estimation problem a Bayesian approach, which allows us to easily include any relevant prior information. Performing inference for such complex model challenging computational as it requires the evaluation intricate high-dimensional integrals. develop here efficient stochastic procedure based on reversible...
In PET image reconstruction, it would be useful to obtain the entire posterior probability distribution of image, because allows for both estimating intensity and assessing uncertainty estimation, thus leading more reliable interpretation. We propose a new entirely probabilistic model: prior is over possible smooth regions (distance-driven Chinese restaurant process), estimated using Gibbs Markov chain Monte Carlo sampler. Data from other modalities (here one or several MR images) are...
This paper considers a problem stemming from the analysis of spectrometric data.When performing experiments on highly radioactive matter, electrical pulses recorded by spectrometer tend to overlap, thus yielding severe distortions when computing histogram pulses' energies.In this paper, we propose fast recursive algorithm which estimates efficiently measurements duration and energies overlapping pulses.Its good performances are shown both simulations real data.Furthermore, its lower...
This work aims at developing a generic virtual source model (VSM) preserving all existing correlations between variables stored in Monte Carlo pre-computed phase space (PS) file, for dose calculation and high-resolution portal image prediction. The reference PS file was calculated using the PENELOPE code, after flattening filter (FF) of an Elekta Synergy 6 MV photon beam. Each particle represented mobile coordinate system by its radial position (r s ) plane, energy (E), polar azimuthal...
We address the problem of X/gamma-ray spectra estimation in fields nuclear physics. Bayesian experimental backgrounds has been studied [1] involving splines. Since Dirichlet processes (DP) sit on discrete measures, they provide an appealing prior for photopeaks. On other hand, order to tackle complexity backgrounds, we consider a Polya Tree Mixture (PTM) - with suitable parameters yielding distribution continuity which predictive densities exhibit better smoothness properties than single...
In this contribution, we propose a discrete-continuous reconstruction method for Positron Emission Tomography (PET). The goal is to reconstruct continuous radiotracer activity distribution from finite set of measurements (namely, the discrete projections detected random emissions). Our approach can be viewed as an indirect density estimation problem, i.e, problem recovering probability function based on observations. We cast in Bayesian nonparametric framework where regularization ill-posed...
The thermal neutron radiative capture cross section for the K isomeric state in $^{177}\mathrm{Lu}$ has been measured first time. Several $^{177}\mathrm{Lu}$${}^{m}$ targets have prepared and irradiated various fluxes at La\"ue Langevin Institute Grenoble CEA reactors OSIRIS ORPHEE Saclay. method consists of measuring $^{178}\mathrm{Lu}$ activity by $\ensuremath{\gamma}$-ray spectroscopy. values obtained four different spectra used to calculate resonance integral $^{177}\mathrm{Lu}$${}^{m}$....
This article presents a problem encountered in nuclear physics, queuing theory and point processes. The studied signal consists of pulses random length energy, possibly sampled, whose time occurrences are points an homogenous Poisson process. Incoming can combine into pile-ups, which results biased estimation the density lengths energies. We introduce model based on two marked processes derive analytical relation between probability function (pdf) observed pile-ups pdf pulses, that leads to...
We introduce a PET reconstruction algorithm following nonparametric Bayesian (NPB) approach. In contrast with expectation maximization (EM), the proposed technique does not rely on any space discretization. Namely, activity distribution - normalized emission intensity of spatial Poisson process is considered as probability density and observations are projections random emissions whose has to be estimated. This approach in sense that quantity interest belongs set measures R <sup...
The ADONIS (Algorithmic Development framewOrk for Nuclear Instrumentation and Spectrometry) system is a new γ spectrometer which addresses high count rate metrology. It has been developed up to 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sup> cps beyond on HPGe detectors. designed in order to: (1) maximize the (pile-up free) output (OCR), (2) achieve both qualitative (i.e. Gaussian shape of spectrum peaks) quantitative reliable...
teaching and research institutions in France or abroad, from public private centers.L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques niveau recherche, publiés ou non, émanant des établissements d'enseignement recherche français étrangers, laboratoires publics privés.
The aim of this work is to propose a method for reconstructing space-time 4D PET images directly from the data without any discretization, neither in space nor time. To accomplish this, we cast reconstruction problem context Bayesian nonparametrics (BNP). activity distribution viewed as an entire probability density on ℝ <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> ×ℝ xmlns:xlink="http://www.w3.org/1999/xlink">+</sup> and inferred...
War against CBRN-E threats needs to continuously develop sensors with improved detection efficiency. More particularly, this topic concerns the NR controls for homeland security. A first analysis requires indeed a fast gamma spectrometry so as detect potential special nuclear materials (SNM). To aim, plastic scintillators could represent best alternative production of large-scale, low-cost radiation portal monitors be deployed on boarders, tolls, etc. Although they are known highly sensitive...
Learning a parametric model from given dataset indeed enables to capture intrinsic dependencies between random variables via conditional probability distribution and in turn predict the value of label variable observed variables. In this paper, we undertake comparative analysis generative discriminative approaches which differ their construction structure underlying inference problem. Our objective is compare ability both leverage information various sources an epistemic uncertainty aware...
Accurate detection of low-level radioactivity is critical for decommissioning projects in nuclear facilities, particularly the design radiation monitoring systems with a low false alarm rate. Utilizing non-parametric Bayesian continuous probability distribution enables reliable mapping potential contamination. Our method introduces statistical test based on Pólya tree prior, applied to detection. The efficiency this proposed quantified using receiver-operating characteristic (ROC) curves and...