Search

International audience

International audience

International audience

Gastrointestinal digestion is a complex dynamic process consisting of mechanically and enzymatically breaking down the food structures into nutrients which are usable by the organism. Monitoring food degradation during the different digestion phases is essential for understanding the mechanisms regulating the digestion of specific nutrients and the impact of food structure on nutrient absorption. In last years, in vitro systems simulating digestion processes [1] have been developped, making it possible to resolve several difficulties associated with in vivo experiments, espetially biological complexity and ethical concerns. MRI is a highly promising non-invasive approach for both in vivo and in vitro digestion research [2] as it can be used to obtain information on the status and amount of water and lipid protons in foods. These information can be used for spatially resolved measurements of multi-scale structural features and composition of food, as demonstrated in recent studies on simplified food such as gelsThe present study aimed to evaluate MRI for monitoring digestion of a complex meal composed of bread, cheese and water. All macronutrients (saccharide, protein, lipid) were gathered in such a meal, with particularities: bread samples containing mainly starch and to a lesser extent (10%) proteins while cheese mainly contained proteins and lipids. Food destructuration associated to erosion and hydrolysis were studied using a semi-dynamic protocol simulating a gastrointestinal human digestive cycle. The study was performed on a 1.5 T MRI scanner (Avanto, Siemens). Evolution of morphological and physico-chemical properties of the particules and the liquid were investigated using: 1) 2D T2 maps 2) 3D fat fraction maps obtained by water-fat separation approach and 3) 3D morphological images obtained by ultra short echo time sequence. Combining different MRI image modalities (Fig. 1), it was possible to investigate separately several phases of the digesta, i. e. supernatant, large food pieces (cheese and bread crust) and molecules or small fragments in suspension which laid deposited at the bottom of the vessel during the MRI measurement.

International audience; One of the top priorities of the aquaculture industry is the genetic improvement of economically important traits in fish, such as those related to processing and quality. However, the accuracy of genetic evaluations has been hindered by a lack of data on such traits from a sufficiently large population of animals. The objectives of this study were thus threefold: (i) to estimate genetic parameters of growth-, yield-, and quality-related traits in rainbow trout (Oncorhynchus mykiss) using three different phenotyping technologies [invasive and non-invasive: microwave-based, digital image analysis, and magnetic resonance imaging (MRI)], (ii) to detect quantitative trait loci (QTLs) associated with these traits, and (iii) to identify candidate genes present within these QTL regions. Our study collected data from 1,379 fish on growth, yield-related traits (body weight, condition coefficient, head yield, carcass yield, headless gutted carcass yield), and quality-related traits (total fat, percentage of fat in subcutaneous adipose tissue, percentage of fat in flesh, flesh colour); genotypic data were then obtained for all fish using the 57K SNP Axiom® Trout Genotyping array. Heritability estimates for most of the 14 traits examined were moderate to strong, varying from 0.12 to 0.67. Most traits were clearly polygenic, but our genome-wide association studies (GWASs) identified two genomic regions on chromosome 8 that explained up to 10% of the genetic variance (cumulative effects of two QTLs) for several traits (weight, condition coefficient, subcutaneous and total fat content, carcass and headless gutted carcass yields). For flesh colour traits, six QTLs explained 1–4% of the genetic variance. Within these regions, we identified several genes (htr1, gnpat, ephx1, bcmo1, and cyp2x) that have been implicated in adipogenesis or carotenoid metabolism, and thus represent good candidates for further functional validation. Finally, of the three techniques used for phenotyping, MRI demonstrated particular promise for measurements of fat content and distribution, while the digital image analysis-based approach was very useful in quantifying colour-related traits. This work provides new insights that may aid the development of commercial breeding programmes in rainbow trout, specifically with regard to the genetic improvement of yield and flesh-quality traits as well as the use of invasive and/or non-invasive technologies to predict such traits.

International audience; The extraction of multi-exponential decay parameters from multi-temporal images corrupted with Rician noise and with limited time samples proves to be a challenging problem frequently encountered in clinical and food MRI studies. This work aims at proposing a method for the estimation of multi-exponential transverse relaxation times from noisy magnitude MRI images. A spatially regularized Maximum-Likelihood estimator accounting for the Rician distribution of the noise is introduced. To deal with the large-scale optimization problem , a Majoration-Minimization approach coupled with an adapted non-linear least squares algorithm is implemented. The proposed algorithm is numerically fast, stable and leads to accurate results. Its effectiveness is illustrated by an application to a simulated phantom and to magnitude multi spin echo MRI images acquired from a tomato sample.

International audience; -Extracting and interpreting multi-exponential relaxation time maps from noisy magnitude MRI images is an ill posed problem that requires solving a large scale inverse problem. A spatially regularized Maximum-Likelihood estimator accounting for the Rician distribution of the noise is introduced. To deal with the large-scale optimization problem, a Majoration-Minimization approach coupled with an adapted non-linear least squares algorithm is implemented. Also, we propose a method for the detection of fruit tissues and their T2 distributions using an unsupervised classification method.; L'obtention puis l'exploitation des cartographies des temps de relaxation multi-exponentielles à l'échelle d'une image entière, à partir des données IRM bruitées de module, nécessite la résolution d'un problème inverse de grande taille. Nous proposons une reconstruction précise des cartographies T2 multi-exponentielles et des intensités relatives associées à l'échelle du voxel. Cette méthode de reconstruction est fondée sur l'estimateur du maximum de vraisemblance exploitant l'hypothèse d'un bruit de Rice et une régularisation spatiale favorisant la régularité des cartographies. Le problème d'optimisation en grande dimension qui en résulte est résolu en utilisant une approche de majoration-minimisation. Enfin, nous proposons une méthode de caractérisation des tissus végétaux à partir des paramètres estimés en utilisant un algorithme de classification non supervisée

International audience; Water status in vegetal cells can be studied using Mutli-exponential NMR relaxation parameters. An original method improving the estimation of the transverse relaxation times from magnitude Magnetic Resonance Images (MRI) has been recently proposed1. In this method, the extraction of the T2 and I0 maps from MRI images, acquired with a multi-spin echo sequence2, is carried out using a spatially regularized optimization algorithm accounting for the Rician noise. K-means clustering is applied to the results in order to regroup voxels with similar T2 and I0 values by classes. To study the distribution of T2 values inside each class, graphs of I0 values in function of their corresponding T2 values are plotted for each class. The main composition of the fruit can be seen, with each class representing a water compartment with different T2 and I0 distributions.

International audience; Synopsis This work aims at improving the estimation of multi-exponential transverse relaxation times from noisy magnitude MRI images. A spatially regularized Maximum-Likelihood estimator accounting for the Rician distribution of the noise was introduced. This approach is compared to a Rician corrected least-square criterion with the introduction of spatial regularization. To deal with the large-scale optimization problem, a majoration-minimization approach was used, allowing the implementation of both the maximum-likelihood estimator and the spatial regularization. The importance of the regularization alongside the rician noise incorporation is shown both visually and numerically on magnitude MRI images acquired on fruit samples. Purpose Multi-exponential relaxation times and their associated amplitudes in an MRI image provide very useful information for assessing the constituents of the imaged sample. Typical examples are the detection of water compartments of plant tissues and the quanti cation of myelin water fraction for multiple sclerosis disease diagnosis. The estimation of the multi-exponential signal model from magnitude MRI images faces the problem of a relatively low signal to noise ratio (SNR), with a Rician distributed noise and a large-scale optimization problem when dealing with the entire image. Actually, maps are composed of coherent regions with smooth variations between neighboring voxels. This study proposes an e cient reconstruction method of values and amplitudes from magnitude images by incorporating this information in order to reduce the noise e ect. The main feature of the method is to use a regularized maximum likelihood estimator derived from a Rician likelihood and a Majorization-Minimization approach coupled with the Levenberg-Marquardt algorithm to solve the large-scale optimization problem. Tests were conducted on apples and the numerical results are given to illustrate the relevance of this method and to discuss its performances. Methods For each voxel of the MRI image, the measured signal at echo time is represented by a multi-exponential model: with The data are subject to an additive Gaussian noise in the complex domain and therefore magnitude MRI data follows a Rician distribution : is the rst kind modi ed Bessel function of order 0 and is the standard deviation of the noise which is usually estimated from the image background. For an MRI image with voxels, the model parameters are usually estimated by minimizing the least-squares (LS) criterion under the assumption of a Gaussian noise using nonlinear LS solvers such as Levenberg-Marquardt (LM). However, this approach does not yield satisfying results when applied to magnitude data. Several solutions to overcome this issue are proposed by adding a correction term to the LS criterion. In this study, the retained correction uses the expectation value of data model under the hypothesis of Rician distribution since it outperforms the other correction strategies: stands for the sum of squares. We refer to this method as Rician corrected LS (RCLS). A more direct way for solving this estimation problem is to use a maximum likelihood (ML) estimator which comes down to minimize: To solve this optimization problem when dealing with the entire image, a majorization-minimization (MM) technique was adopted. The resulting MM-ML algorithm is summarized in gure 1, the LM algorithm used in this method minimizes a set of LS criteria derived from the quadratic majorization strategy. A spatial regularization term based on a cost function was also added to both criteria (and) to ensure spatial smoothness of the estimated maps. In order to reduce the numerical complexity by maintaining variable separability between each voxel and it's neighboring voxels , the function is majorized by : where stands for the iteration number of the iterative optimization algorithm.

High-throughput phenotyping of shellfish’s traits constitutes a challenge due to the specificities of aquatic animals. Appropriate measuring tools have often to be developed or adapted, in order to continuously and automatically acquire data of biomarkers of interest. IMAGIGAS project aimed to explore a non-destructive and non-invasive approach by using magnetic resonance imaging (MRI) to monitor at high-throughput two biomarkers (body weight and temperature) of the pacific oyster, before and during infectious challenges by Ostreid Herpesvirus type 1 (OsHV-1). Pacific oysters and OsHV-1 suspension and contaminated seawater were produced at Ifremer facilities. The experiments were conducted in 2016-2017 by three staffs (Ifremer LGPMM, Irstea IRM Food, Université de La Rochelle LIENSs) at PRISM platform of Irstea at Rennes who provided a Siemens Avanto 1,5 T MRI. One objective was to monitor body weight of numerous individuals. Protocols were inspired by previous Ifremer / Irstea collaborative works and allowed to increase the throughput of measure to 69 oysters every 45 minutes. The estimation of body weight is based on a relation established between the sum of pixels whose grey level is greater than a particular value in pictures taken by MRI and the weight of drained then lyophilized flesh of oysters measured by precision balance. Each development experiments were completed by application experiments with infection of oysters by OsHV-1 and measures of body weight.

This paper presents a method to characterize the development of the structures of puff pastries during proving using MRI. Since the resolution is large, each pixel contains an unknown proportion of three components: fat, paste and gas. The signal to noise ratio is low since gas which reaches 80% at the end of proving gives no signal. The signal is the sum of reference signals, corresponding to pixels filled with one component, weighted by the proportions. The reference signals were supposed to be known. We adopted an edge-preserving approach based on the minimization of a penalized least-square criterion. This criterion is the weighted sum of a term accounting for the fidelity to data and a regularization term. The minimization of the criterion is based on a non-linear conjugate gradient algorithm. The settings of the weights of the two terms is based on simulations. Then simulation results are presented. The mean error was similar with or without regularization and depended on the components and their proportion (less that 1% up to 6%). Fat and gas proportions were overestimated, paste proportion was underestimated. The dispersion of the results was lower with regularization (from 0.3% up to 1.5 %). Monte-Carlo simulations showed that these results were not influenced very much by the uncertainty on the reference signals at the end of the proving. Larger uncertainties were found at the beginning of proving. We showed that the regularization of the solutions did improve the visualization of the structures confirming the interest of this approach.We also found that layers down to 40 microns thick and bubbles of which size exceeded 2.5mm could be easily distinguished in fat and gas proportion maps respectively. Experiments on genuine MRI images of Danish paste confirmed the results obtained with the simulation study. We were able to validate the method at the scale of the Danish paste by observing the evolution of the sum of the fat proportion over time, as well as the evolution of the gas content compared with the evolution of the size of the pastry. This confirmed the possibility to use the method to study the proving of a Danish paste.

In magnetic resonance imaging, it is usual to perform several repeated acquisitions in order to reduce the impact of the noise on the evaluation of the average signal intensity. The main goal of this paper is to examine whether it is preferable to consider the magnitude of each complex measurement, or the single magnitude of the averaged data at each voxel. A thorough information theoretic study is proposed, which shows that the second option is preferable in all cases. In order to reach such a conclusion, several properties of the Fisher information of Rician distributions are proved, the main mathematical result being that it is an increasing function of the signal intensity, at constant noise level, when the latter is known. The case of an unknown noise level is also considered, with essentially the same conclusion. We also propose a more empirical comparison involving maximum likelihood estimation from simulated data, which is in agreement with the information theoretic analysis. Finally, an extension to imaging systems where multiple-receiver coils are employed is considered.

Genetic parameters of lipids heterogeneity estimated by direct (MRI) or indirect (artificial vision, Fat meter) technologies in the fillet of large rainbow trout. Aquaculture conference: To the Next 40 Years of Sustainable Global Aquaculture

International audience; Introduction Fish selection for/or against fillet fatness required sib selection to measure lipids in the fillets or alternative indirect technologies usable on live candidates. Genetic parameters of different fillet adiposity components were evaluated by MRI (total fat; dorsal or ventral subcutaneous fat representing different trimming losses), artificial vision with A3 scanner (myosepta surfaces, dorsal or ventral or subcutaneous fat) and Fat Meter (external indirect measure applicable on live fish) technologies to optimize sib or individual selection on fillet adiposity. Methods 1625 rainbow trout from a partial factorial mating design of 90 dams x 100 sires (900 families) were slaughtered at 18 months old (1580g), processed and pedigree assigned with microsatellites. Genetic parameters were estimated with animal model (VCE). Results Heritabilities of myosepta (0.15-0.29) and adipose tissues surfaces (0.37-0.43) estimated with artificial vision are intermediates, as those of Fat Meter measures (0.38-0.45). Heritabilities of the total fat in the fillet or of the subcutaneous fat reported to the total surface of the cutlet or the dorsal or ventral related surfaces estimated with MRI were highest (h2> 0.60). The genetic correlations between total fat and the relative surface of subcutaneous fat are intermediate (0.64) and higher between the Fat Meter measure and the relative subcutaneous fat surface (0.77) than the total fat (0.43). Discussion All components of the fillet adiposity exhibited intermediate heritability. The two time lower heritability of the Fat Meter measure in this experiment and its intermediate genetic correlation with the total fat of the fillet is making selection on this indirect trait less efficient than selection on sibs. The high genetic correlation with the relative subcutaneous fat informed that the Fat Meter estimates mostly this later trait in large rainbow trout. MRI and artificial vision technologies are then recommended for more efficient selection programs in large fish to manipulate filet adiposity of trimming yields.

National audience; L'agroalimentaire et le génie civil constituent deux domaines d'applications très particuliers pour l'imagerie par résonance magnétique (IRM). Essentiellement axées sur les problèmes d'élaboration, de conditionnement et d'évolution de produits et de matériaux en contexte industriel, les études doivent souvent être menées sur des échantillons massifs. Elles mettent en oeuvre des imageurs de grande taille et réclament le développement de méthodologies spécifiques par rapport aux applications médicales. Des systèmes originaux d'instrumentation et de sollicitation in situ des matériaux ont pu être conçus dans ce cadre, nourris par des collaborations étroites entre différents corps de métiers. Deux laboratoires, l'Unité Technologie des équipements agroalimentaires de l'Irstea et le Laboratoire Navier de l'Université Paris-Est, livrent leur témoignage, fruit de plusieurs années d'expertise. Food-processing and civil engineering are very particular application fields for magnetic resonance imaging (MRI). Studies mainly focus on preparation, packaging, and further evolution of products and materials in industry. They are often carried out on massive samples. They involve the use of large MRI facilities, and require the development of specific methodology as compared with routine medical studies. Stimulating close collaborations between various technical specialties, they saw the raise of original experimental setups for in situ measurements and sample solicitations. Two French laboratories, the " Technologie des équipements agroalimentaires " Unity at Irstea and " Laboratoire Navier " at Paris-Est University, report on their long standing expertise.

Le but du projet était d’évaluer chez la truite arc-en-ciel la faisabilité du dosage de la teneur en lipides intramusculaires et de la surface de gras sous cutané par Imagerie en Résonance Magnétique (IRM) sur des darnes entières après décongélation. La possibilité d’analyser plusieurs darnes simultanément dans l’IRM a été validée en utilisant un support spécifique pour disposer les darnes au sein de l’imageur. Un modèle de correction du signal permet de pallier l’inhomogénéité du signal dans l’espace de mesure. La comparaison de la teneur en lipides de darnes fraîches et décongelées avec un dosage RMN met en évidence de très bonnes corrélations avec une erreur moyenne de 0.8 et 0.89% respectivement. Ce qui démontre que la congélation ne dégrade pas la précision du dosage. La surface de gras sous cutané mesurée en IRM et en vision numérique présente de bonnes corrélations R²=0.77 et 0.87 respectivement pour les tissus adipeux ventraux et dorsaux.

absent

Magnetic resonance imaging (MRI) is a non-invasive modality designed for clinical diagnosis. Other domains also exploit this technique, such as food products analysis. The applicative aim of our work is the study of the repartition of fat tissues in fish. We are particularly interested in the quantification of the tissues. An accurate quantification requires the denoising of the images and the correction of the intensity inhomogeneities due to the spatial variation of the radiofrequency magnetic field (RF). We use T1-weighted images. In this case, the effects of the RF inhomogeneities are complex since the bias that is induced in the images depends on the tissue. The proposed method takes place in the inverse problem framework where denoising and correcting are tackled jointly. It is based on a physical model of the MRI signal and a model of the sample considered as made of a finite number of tissues. The method relies on the minimisation of a penalised criterion. This criterion consists of a data-fitting term added with regularisation terms in order to ensure spatially smooth solutions while preserving the edges in the image. The method needs several images acquired with different protocols. The minimisation is based on a block-coordinate descent approach where each block consists in iterations of the conjugate gradient algorithm. Results obtained on images of fish validate our approach. We also present preliminary results on the optimisation of the choice of the protocols which lead to the best estimation of the variables. These results rely on the theory of experiment planning.; L'imagerie par résonance magnétique (IRM) est une modalité non-invasive développée pour le diagnostic clinique. D'autres domaines se sont approprié cette technique, comme l'analyse de produits agroalimentaires. Le cadre applicatif de nos travaux est l'étude de la répartition des tissus adipeux chez le poisson en IRM bas champ. Au-delà de la visualisation, c'est la quantification des tissus qui nous intéresse ici. Une quantification précise requiert le débruitage des images et la correction des inhomogénéités d'intensité liées à la variation spatiale du champ magnétique radiofréquence (RF). En IRM pondérée-T1 utilisée ici, les inhomogénéités de la RF ont un effet complexe et introduisent un biais qui dépend du tissu en présence. La méthode proposée aborde de façon unifiée la correction et le débruitage dans le cadre de la résolution des problèmes inverses. Elle prend en compte un modèle de biais issu de la physique de l'IRM auquel s'ajoute un modèle de l'échantillon vu comme une somme pondérée de tissus. La méthode est basée sur la minimisation d'un critère pénalisé comprenant des termes d'attache aux données et des termes de régularisation assurant des solutions spatialement lisses tout en conservant les contours dans l'image. Elle impose d'acquérir plusieurs images avec des protocoles différents. La minimisation est basée sur une résolution par blocs de variables, chaque bloc faisant appel à l'algorithme du gradient conjugué. Des résultats obtenus sur des images de poisson valident l'approche. Nous présentons de plus les résultats préliminaires d'une démarche de planification d'expérience pour choisir les protocoles permettant une estimation optimale des variables.

International audience; Nous proposons une méthode pour réduire le bruit dans les images IRM et éliminer les effets des inhomogénéités spatiales des antennes d'émission et de réception de la radiofréquence (RF). Nous traitons particulièrement le cas des images pondérées T1 acquises en IRM bas champ. Le biais engendré par l'inhomogénéité spatiale dela RF en émission dépend du tissu imagé. Il ne peut être considéré comme purement multiplicatif et indépendant du contenu de l'image. Afin de le corriger, nous considérons que chaque voxel contient un mélange des tissus susceptibles d'être présents dans l'image. Le signal IRM est modélisé par la somme pondérée des signaux de chaque tissu. Le nombre de tissus ainsi que certaines caractéristiques RMN des tissus sont supposés connus. Plusieurs images, acquises avec des paramètres d'acquisition différents, sont nécessaires. Un critère des moindres carrés pénalisé est utilisé afin d'estimer les inhomogénéités de la RF en émission, en réception ainsi que la proportion des tissus. La minimisation de ce critère est réalisée à l'aide du gradient conjugué utilisé dans un schéma itératif de type Gauss-Seidel par blocs. Des résultats obtenus sur des images simulées puis réelles sont présentés.

National audience