Carole Frindel

Towards In Vivo Diffusion Tensor MRI on Human Heart using Edge-Preserving Regularization

We investigate the noise sensitivity in various diffusion tensor MRI acquisition protocols in sixteen human ex vivo hearts. In particular, we compare the accuracy of protocols with various numbers of excitations and diffusion sensitizing directions for estimating the principal diffusion directions in the myocardium. It is observed that noise sensitivity decreases as the number of excitations and the number of sensitizing directions increase (and hence as the acquisition time increases). To reduce the effects of noise and to improve the results obtained with a smaller number of excitations and/or a smaller number of sensitizing directions, we introduce a 3-D edge-preserving regularization method operating on diffusion weighted images. It allows to maintain the quality of the principal diffusion direction field while minimizing the acquisition time, which is a necessary step for in vivo diffusion tensor MR imaging of the human heart.

A Sensitive and Automatic White Matter Fiber Tracts Model for Longitudinal Analysis of Diffusion Tensor Images in Multiple Sclerosis.

Diffusion tensor imaging (DTI) is a sensitive tool for the assessment of microstructural alterations in brain white matter (WM). We propose a new processing technique to detect, local and global longitudinal changes of diffusivity metrics, in homologous regions along WM fiber-bundles. To this end, a reliable and automatic processing pipeline was developed in three steps: 1) co-registration and diffusion metrics computation, 2) tractography, bundle extraction and processing, and 3) longitudinal fiber-bundle analysis. The last step was based on an original Gaussian mixture model providing a fine analysis of fiber-bundle cross-sections, and allowing a sensitive detection of longitudinal changes along fibers. This method was tested on simulated and clinical data. High levels of F-Measure were obtained on simulated data. Experiments on cortico-spinal tract and inferior fronto-occipital fasciculi of five patients with Multiple Sclerosis (MS) included in a weekly follow-up protocol highlighted the greater sensitivity of this fiber scale approach to detect small longitudinal alterations.

A 3-D spatio-temporal deconvolution approach for MR perfusion in the brain

We propose an original spatio-temporal deconvolution approach for perfusion-weighted MRI applied to cerebral ischemia. The regularization of the underlying inverse problem is achieved with spatio-temporal priors and the resulting optimization problem is solved by half-quadratic minimization. Our approach offers strong convergence guarantees, including when the spatial priors are non-convex. Moreover, experiments on synthetic data and on real data collected from subjects with ischemic stroke show significant performance improvements over the standard approaches-namely, temporal deconvolution based on either truncated singular-value decomposition or ℓ2-regularization-in terms of various performance measures.

Validity of Shape as a Predictive Biomarker of Final Infarct Volume in Acute Ischemic Stroke

Background and Purpose— This study examines whether lesion shape documented on magnetic resonance diffusion-weighted imaging during acute stroke improves the prediction of the final infarct volume compared with lesion volume only. Methods— Diffusion-weighted imaging data and clinical information were retrospectively reviewed in 110 consecutive patients who underwent (n=67) or not (n=43) thrombolytic therapy for acute ischemic stroke. Three-dimensional shape analysis was performed on admission diffusion-weighted imaging data and 5 shape descriptors were developed. Final infarct volume was measured on T2-fluid-attenuated inversion recovery imaging data performed 30 days after stroke. Results— Shape analysis of acute ischemic lesion and more specifically the ratio of the bounding box volume to the lesion volume before thrombolytic treatment improved the prediction of the final infarct for patients undergoing thrombolysis (R2=0.86 in model with volume; R2=0.98 in model with volume and shape). Conclusions— Our findings suggest that lesion shape contains important predictive information and reflects important environmental factors that might determine the progression of ischemia from the core.

Computer vision tools to optimize reconstruction parameters in x-ray in-line phase tomography

In this article, a set of three computer vision tools, including scale invariant feature transform (SIFT), a measure of focus, and a measure based on tractography are demonstrated to be useful in replacing the eye of the expert in the optimization of the reconstruction parameters in x-ray in-line phase tomography. We demonstrate how these computer vision tools can be used to inject priors on the shape and scale of the object to be reconstructed. This is illustrated with the Paganin single intensity image phase retrieval algorithm in heterogeneous soft tissues of biomedical interest, where the selection of the reconstruction parameters was previously made from visual inspection or physical assumptions on the composition of the sample.

Information-based analysis of X-ray in-line phase tomography with application to the detection of iron oxide nanoparticles in the brain

The study analyzes noise in X-ray in-line phase tomography in a biomedical context. The impact of noise on detection of iron oxide nanoparticles in mouse brain is assessed. The part of the noise due to the imaging system and the part due to biology are quantitatively expressed in a Neyman Pearson detection strategy with two models of noise. This represents a practical extension of previous work on noise in phase-contrast X-ray imaging which focused on the theoretical expression of the signal-to-noise ratio in mono-dimensional phantoms, taking account of the statistical noise of the imaging system only. We also report the impact of the phase retrieval step on detection performance. Taken together, this constitutes a general methodology of practical interest for quantitative extraction of information from X-ray in-line phase tomography, and is also relevant to assessment of contrast agents with a blob-like signature in high resolution imaging.

A global approach to cardiac tractography

Cardiac myofibrilles bundles in the human heart can be located by iteratively tracing the local water diffusion direction inferred from diffusion weighted MRI images. This well known process, called streamlining, needs to be initialized by a seed. In this paper, a global, graph modeling approach for cardiac myofibrille tractography is presented. Seed points are no longer needed : it predicts the fibers in one shot for the whole DT-MRI volume without initialization artifacts. The main merits of the presented methodology are its ability to give a unique estimation of the heart architecture (independent of seed initialization), the fact that it has no hyper- parameters and that it provides an optimal balance between the density of fibers and the amount of available data.

A longitudinal model for variations detection in white matter fiber-bundles

Processing of longitudinal diffusion tensor imaging (DTI) data is a crucial challenge to better understand pathological mechanisms of complex brain diseases such as multiple sclerosis (MS) where white matter (WM) fiber bundles are variably altered by inflammatory events. In this work, we propose a new fully automated method to detect significant longitudinal changes in diffusivity metrics along WM fiber-bundles. This method consists in two steps: i) preprocessing of longitudinal diffusion acquisitions and WM fiber-bundles extraction, ii) application of a genetic algorithm (GA) to detect “pathological” changes. This method was applied first, on simulated longitudinal variations, and second, on MS patients longitudinal data. High level of precision, recall and F-Measure were obtained for the detection of small longitudinal changes along the WM fiber-bundles.

Cardiac Fibre Trace Clustering for the Interpretation of the Human Heart Architecture

Cardiac fibre architecture plays a key role in heart function. Recently, the estimation of fibre structure has been simplified with diffusion tensor MRI (DT-MRI). In order to assess the heart architecture and its underlying function, with the goal of dealing with pathological tissues and easing inter-patient comparisons, we propose a methodology for finding cardiac myofibrille trace correspondences across a fibre population obtained from DT-MRI data. It relies on the comparison of geometrical and topological clustering operating on different fibre representation modes (fixed length sequences of 3-D coordinates with or without ordering strategy, and 9-D vectors for trace shape approximation). In geometrical clustering (or k-means) each fibre path is assigned to the cluster with nearest barycenter. In topological (or spectral) clustering the data is represented by a similarity graph and the graph vertices are divided into groups so that intra-cluster connectivity is maximized and inter-cluster connectivity is minimized. Using these different clustering methods and fibre representation modes, we predict different fibre trace classifications for the same cardiac dataset. These classification results are compared to the human heart architecture models proposed in the literature.

Image processing for precise three-dimensional registration and stitching of thick high-resolution laser-scanning microscopy image stacks

The possible depth of imaging of laser-scanning microscopy is limited not only by the working distances of objective lenses but also by image degradation caused by attenuation and diffraction of light passing through the specimen. To tackle this problem, one can either flip the sample to record images from both sides of the specimen or consecutively cut off shallow parts of the sample after taking serial images of certain thickness. Multiple image substacks acquired in these ways should be combined afterwards to generate a single stack. However, subtle movements of samples during image acquisition cause mismatch not only in the translation along x-, y-, and z-axes and rotation around z-axis but also tilting around x- and y-axes, making it difficult to register the substacks precisely. In this work, we developed a novel approach called 2D-SIFT-in-3D-Space using Scale Invariant Feature Transform (SIFT) to achieve robust three-dimensional matching of image substacks. Our method registers the substacks by separately fixing translation and rotation along x-, y-, and z-axes, through extraction and matching of stable features across two-dimensional sections of the 3D stacks. To validate the quality of registration, we developed a simulator of laser-scanning microscopy images to generate a virtual stack in which noise levels and rotation angles are controlled with known parameters. We illustrate quantitatively the performance of our approach by registering an entire brain of Drosophila melanogaster consisting of 800 sections. Our approach is also demonstrated to be extendable to other types of data that share large dimensions and need of fine registration of multiple image substacks. This method is implemented in Java and distributed as ImageJ/Fiji plugin. The source code is available via Github (http://www.creatis.insa-lyon.fr/site7/fr/MicroTools).