This paper investigates a new method for cardiac motion estimation in 2D ultrasound images. The motion estimation problem is formulated as an energy minimization with spatial and sparse regularizations. In addition to a classical spatial smoothness constraint, the proposed method exploits the sparse properties of the cardiac motion to regularize the solution via an appropriate dictionary learning step. The proposed method is evaluated in terms of motion estimation and strain accuracy and compared with state-of-the-art algorithms using a dataset of realistic simulations. These simulation results show that the proposed method provides very promising results for myocardial motion estimation.

Les composants électroniques destinés à l’échantillonnage haute fréquence des signaux ont des temps de réponse susceptibles de dégrader les performances des algorithmes de reconstruction. Dans cet article, on considère un signal échantillonné par un composant modélisé par un filtre linéaire invariant activé à des instants potentiellement irréguliers et qui forment une suite vérifiant la condition de Landau (généralisation de la condition de Nyquist). On étudie les conséquences des imperfections des échantillonneurs dans la reconstruction du signal et dans l’estimation de l’enveloppe et de la phase.

Dans cet article, nous proposons d’étudier différents algorithmes d’égalisation pour signaux à Modulation de Phase Continue (continuous phase modulation, CPM) dans le domaine fréquentiel à l’aide d’une représentation polyphase du signal. Par cette approche unifiée, nous montrons l’équivalence analytique de ces égaliseurs sous certaines hypothèses liées au canal de transmission, les liens, ainsi que leurs limitations dans certains cas.

Cardiac strain estimation from ultrasound images is an efficient tool for the diagnosis of cardiac diseases. This study focuses on cardiac amyloidosis, a pathology characterized by non-specific early symptoms such as the increased wall thickness. Recent clinical studies have demonstrated that patients with cardiac amyloidosis present an apex-to-base gradient longitudinal strain pattern, i.e., a normal strain in apex and abnormally lower values for base segments. Existing cardiac motion estimation methods belong to three categories based on optical flow, speckle tracking and elastic registration. To overcome the ill-posedness of motion estimation, they use local parametric models (e.g., affine) or global regularizations (e.g., B-splines). The objective of this study is to evaluate a recently proposed cardiac motion estimation method based on dictionary learning on patients subjected to cardiac amyloidosis.

Le démélange d’images hyperspectrales est un exemple particulier du problème de factorisation en matrices non-négatives (NMF) qui consiste à identifier les signatures spectrales d’un milieu imagé ainsi que leurs proportions dans chacun des pixels. Toutefois, le nombre important de pixels composant ces images peut s’avérer contraignant en termes de mémoire ou de temps de calcul, ce qui peut motiver le développement de techniques d’estimation distribuées (entre plusieurs processeurs et/ou plusieurs machines). Dans cette perspective, nous proposons une approche distribuée du problème de démélange, basée sur de récentes avancées en optimisation distribuée asynchrone inspirées de l’algorithme proximal alternating linearized minimization (PALM). L’intérêt d’une estimation asynchrone par rapport à une procédure synchrone est illustré dans ce contexte sur des données synthétiques.

A classical problem in hyperspectral imaging, referred to as hyperspectral unmixing, consists in estimating spectra associated with each material present in an image and their proportions in each pixel. In practice, illumination variations (e.g., due to declivity or complex interactions with the observed materials) and the possible presence of outliers can result in significant changes in both the shape and the amplitude of the measurements, thus modifying the extracted signatures. In this context, sequences of hyperspectral images are expected to be simultaneously affected by such phenomena when acquired on the same area at different time instants. Thus, we propose a hierarchical Bayesian model to simultaneously account for smooth and abrupt spectral variations affecting a set of multitemporal hyperspectral images to be jointly unmixed. This model assumes that smooth variations can be interpreted as the result of endmember variability, whereas abrupt variations are due to significant changes in the imaged scene (e.g., presence of outliers, additional endmembers, etc.). The parameters of this Bayesian model are estimated using samples generated by a Gibbs sampler according to its posterior. Performance assessment is conducted on synthetic data in comparison with state-of-the-art unmixing methods.

This paper studies a new Bayesian algorithm for the joint reconstruction and classification of reflectance confocal microscopy (RCM) images, with application to the identification of human skin lentigo. The proposed Bayesian approach takes advantage of the distribution of the multiplicative speckle noise affecting the true reflectivity of these images and of appropriate priors for the unknown model parameters. A Markov chain Monte Carlo (MCMC) algorithm is proposed to jointly estimate the model parameters and the image of true reflectivity while classifying images according to the distribution of their reflectivity. Precisely, a Metropolis-within-Gibbs sampler is investigated to sample the posterior distribution of the Bayesian model associated with RCM images and to build estimators of its parameters, including labels indicating the class of each RCM image. The resulting algorithm is applied to synthetic data and to real images from a clinical study containing healthy and lentigo patient.