A5060957707- Statistical Methods and Inference
- Advanced Clustering Algorithms Research
- Advanced Statistical Methods and Models
- Bayesian Methods and Mixture Models
- Data Management and Algorithms
- Data Mining Algorithms and Applications
- Neural Networks and Applications
- Bayesian Modeling and Causal Inference
- Multiple Sclerosis Research Studies
- Advanced Statistical Process Monitoring
- Odor and Emission Control Technologies
- Hydrological Forecasting Using AI
- Cardiovascular Function and Risk Factors
- Gaussian Processes and Bayesian Inference
- Machine Learning in Healthcare
- Time Series Analysis and Forecasting
- Advanced MRI Techniques and Applications
- Schizophrenia research and treatment
- Glioma Diagnosis and Treatment
- Oil Palm Production and Sustainability
- Cardiomyopathy and Myosin Studies
- Face and Expression Recognition
- Aquatic Ecosystems and Phytoplankton Dynamics
- Advanced Text Analysis Techniques
- Algorithms and Data Compression
Aix-Marseille Université
2015-2025
Institut de Mathématiques de Marseille
2014-2024
Institut Polytechnique de Bordeaux
2000-2024
Château Gombert
1999-2024
Centre National de la Recherche Scientifique
2004-2023
Centrale Marseille
2017-2022
Institut Fresnel
2022
Shaikh Khalifa Medical City
2022
Laboratoire de Mathématiques d'Orsay
2011-2016
Creativ-Ceutical (France)
2014
<h3>BACKGROUND AND PURPOSE:</h3> Pretherapeutic determination of tumor grade and genotype in II III oligodendroglial tumors is clinically important but still challenging. Tumor 1p/19q status are currently the 2 most factors therapeutic decision making for patients with these tumors. Histopathology cMRI studies limited some cases. In present study, we were interested determining whether combination PWI, DWI, MR spectroscopy could help distinguish according to their histopathologic genotype....
Background: Previous studies have shown an antiasthenic effect of citrulline/malate (CM) but the mechanism action at muscular level remains unknown. Objective: To investigate effects CM supplementation on muscle energetics. Methods: Eighteen men complaining fatigue with no documented disease were included in study. A rest-exercise (finger flexions)-recovery protocol was performed twice before (D−7 and D0), three times during (D3, D8, D15), once after (D22) 15 days oral 6 g/day CM. Metabolism...
Background Facioscapulohumeral muscular dystrophy type 1 (FSHD1) is the third most common inherited dystrophy. Considering highly variable clinical expression and slow disease progression, sensitive outcome measures would be of interest. Methods Findings Using muscle MRI, we assessed fatty infiltration in lower limbs 35 FSHD1 patients 22 healthy volunteers by two methods: a quantitative imaging (qMRI) combined with dedicated automated segmentation method performed on both thighs standard...
Clustering is widely used in unsupervised learning to find homogeneous groups of observations within a dataset. However, clustering mixed‐type data remains challenge, as few existing approaches are suited for this task. This study presents the state‐of‐the‐art these and compares them using various simulation models. The compared methods include distance‐based k ‐prototypes, PDQ, convex ‐means, probabilistic KAy‐means MIxed LArge (KAMILA), mixture Bayesian networks (MBNs), latent class model...
Exertional heat stroke (EHS) is usually triggered by strenuous exercise performed under hot and humid environmental conditions. Although the pathogenesis of an EHS episode differs from that a clinical malignant hyperthermia (MH) crisis, both conditions share some similarities in symptoms, such as abnormal increase core temperature. By use 31P magnetic resonance spectroscopy, we analyzed muscle energetics 26 post-EHS subjects for whom vitro halothane/caffeine contracture tests were...
Multiple sclerosis (MS), a common neurodegenerative disease, has well-described associations with quality of life (QoL) impairment. QoL changes found in longitudinal studies are difficult to interpret due the potential response shift (RS) corresponding respondents' changing standards, values, and conceptualization QoL. This study proposes test capacity Random Forest (RF) for detecting RS reprioritization as relative importance domains' over time. was observational study. The main inclusion...
Prolonged Hospital Length of Stay (PLOS) is an indicator deteriorated efficiency in Quality Care. One goal public health management to reduce PLOS by identifying its most relevant predictors. The objective this study explore Machine Learning (ML) models that best predict PLOS.Our dataset was collected from the French Medico-Administrative database (PMSI) as a retrospective cohort all discharges year 2015 large university hospital France (APHM). outcomes were LOS transformed into binary...
In the Design of Experiments, we seek to relate response variables explanatory factors. Response Surface methodology (RSM) approximates relation between output and a polynomial transform using linear model. Some researchers have tried adjust other types models, mainly nonlinear nonparametric. We present large panel Machine Learning approaches that may be good alternatives classical RSM approximation. The state art such is given, including classification regression trees, ensemble methods,...
Applying automated segmentation to UK Biobank MRI scans, hypertension, higher body mass index, and physical activity levels were associated with increased left ventricle trabeculations in healthy middle-aged White adults.Key Results:1.In a cross-sectional secondary analysis of prospectively collected data from the Biobank, applying cardiac trabeculated ventricular (LV) was significantly adults established cardiovascular risk factors (n = 28 672) compared reference group 7384) for both men...
The aim of this study (EPIDIAB) was to assess the relationship between epicardial adipose tissue (EAT) and micro macrovascular complications (MVC) type 2 diabetes (T2D). EPIDIAB is a post hoc analysis from AngioSafe T2D study, which multicentric aimed at determining safety antihyperglycemic drugs on retina including patients with screened for diabetic retinopathy (DR) (n = 7200) deeply phenotyped MVC. Patients included who had undergone cardiac CT CAC (Coronary Artery Calcium) scoring after...
Various bioinformatics analyses provide sets of genomic coordinates interest. Whether two such possess a functional relation is frequent question. This often determined by interpreting the statistical significance their overlaps. However, only few existing methods consider lengths overlap, and they do not resolutive p-value.
Background. Interpretation of quality life (QoL) scores over time can be difficult because possible changes in internal standards, values, and conceptualization QoL by individuals. This effect is called a response shift (RS). The purpose this study was to examine whether an RS occurred 24-mo period patients who were suffering from schizophrenia. Methods. random forest method applied detect any reprioritization multicenter cohort study. recorded using generic questionnaire (SF36) at baseline...
With the strengths of deep learning, computer-aided diagnosis (CAD) is a hot topic for researchers in medical image analysis. One main requirements training learning model providing enough data network. However, images, due to difficulties collection and privacy, finding an appropriate dataset (balanced, samples, etc.) quite challenge. Although synthesis could be beneficial overcome this issue, synthesizing 3D images hard task. The objective paper generate T1 weighted MRI corresponding...
The termination of mature eukaryotic mRNAs occurs at specific polyadenylation sites located downstream from stop codons in the 3'-untranslated region (UTR). An accurate delineation these is essential for study 3'-UTR-based gene regulation and design pertinent probes transcriptome analysis. Although typical poly(A) are between 0 2 kb codon, EST sequence analyses have identified unexpectedly long ranges (5-10 kb) a number genes. Here we perform complete mapping full-length cDNA sequences on...
The aim was to develop a multidimensional computerized adaptive short-form questionnaire, the MusiQoL-MCAT, from fixed-length QoL questionnaire for multiple sclerosis. A total of 1992 patients were enrolled in this international cross-sectional study. development MusiQoL-MCAT based on assessment between-items MIRT model fit followed by real-data simulations. MCAT algorithm Bayesian maximum posteriori estimation latent traits and Kullback–Leibler information item selection. We examined...