Showing papers by "Nicholas Ayache published in 2015"

The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)

Abstract: In this paper we report the set-up and results of the Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) organized in conjunction with the MICCAI 2012 and 2013 conferences Twenty state-of-the-art tumor segmentation algorithms were applied to a set of 65 multi-contrast MR scans of low- and high-grade glioma patients—manually annotated by up to four raters—and to 65 comparable scans generated using tumor image simulation software Quantitative evaluations revealed considerable disagreement between the human raters in segmenting various tumor sub-regions (Dice scores in the range 74%–85%), illustrating the difficulty of this task We found that different algorithms worked best for different sub-regions (reaching performance comparable to human inter-rater variability), but that no single algorithm ranked in the top for all sub-regions simultaneously Fusing several good algorithms using a hierarchical majority vote yielded segmentations that consistently ranked above all individual algorithms, indicating remaining opportunities for further methodological improvements The BRATS image data and manual annotations continue to be publicly available through an online evaluation system as an ongoing benchmarking resource

Benchmark for Algorithms Segmenting the Left Atrium From 3D CT and MRI Datasets

Abstract: Knowledge of left atrial (LA) anatomy is important for atrial fibrillation ablation guidance, fibrosis quantification and biophysical modelling. Segmentation of the LA from Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) images is a complex problem. This manuscript presents a benchmark to evaluate algorithms that address LA segmentation. The datasets, ground truth and evaluation code have been made publicly available through the http://www.cardiacatlas.org website. This manuscript also reports the results of the Left Atrial Segmentation Challenge (LASC) carried out at the STACOM’13 workshop, in conjunction with MICCAI’13. Thirty CT and 30 MRI datasets were provided to participants for segmentation. Each participant segmented the LA including a short part of the LA appendage trunk and proximal sections of the pulmonary veins (PVs). We present results for nine algorithms for CT and eight algorithms for MRI. Results showed that methodologies combining statistical models with region growing approaches were the most appropriate to handle the proposed task. The ground truth and automatic segmentations were standardised to reduce the influence of inconsistently defined regions (e.g., mitral plane, PVs end points, LA appendage). This standardisation framework, which is a contribution of this work, can be used to label and further analyse anatomical regions of the LA. By performing the standardisation directly on the left atrial surface, we can process multiple input data, including meshes exported from different electroanatomical mapping systems.

Efficient Lattice Boltzmann Solver for Patient-Specific Radiofrequency Ablation of Hepatic Tumors

Abstract: Radiofrequency ablation (RFA) is an established treatment for liver cancer when resection is not possible. Yet, its optimal delivery is challenged by the presence of large blood vessels and the time-varying thermal conductivity of biological tissue. Incomplete treatment and an increased risk of recurrence are therefore common. A tool that would enable the accurate planning of RFA is hence necessary. This manuscript describes a new method to compute the extent of ablation required based on the Lattice Boltzmann Method (LBM) and patient-specific, pre-operative images. A detailed anatomical model of the liver is obtained from volumetric images. Then a computational model of heat diffusion, cellular necrosis, and blood flow through the vessels and liver is employed to compute the extent of ablated tissue given the probe location, ablation duration and biological parameters. The model was verified against an analytical solution, showing good fidelity. We also evaluated the predictive power of the proposed framework on ten patients who underwent RFA, for whom pre- and post-operative images were available. Comparisons between the computed ablation extent and ground truth, as observed in postoperative images, were promising (DICE index: 42%, sensitivity: 67%, positive predictive value: 38%). The importance of considering liver perfusion while simulating electrical-heating ablation was also highlighted. Implemented on graphics processing units (GPU), our method simulates 1 minute of ablation in 1.14 minutes, allowing near real-time computation.

Bayesian personalization of brain tumor growth model

Abstract: Recent work on brain tumor growth modeling for glioblastoma using reaction-diffusion equations suggests that the diffusion coefficient and the proliferation rate can be related to clinically relevant information. However, estimating these parameters is difficult due to the lack of identifiability of the parameters, the uncertainty in the tumor segmentations, and the model approximation, which cannot perfectly capture the dynamics of the tumor. Therefore, we propose a method for conducting the Bayesian personalization of the tumor growth model parameters. Our approach estimates the posterior probability of the parameters, and allows the analysis of the parameters correlations and uncertainty. Moreover, this method provides a way to compute the evidence of a model, which is a mathematically sound way of assessing the validity of different model hypotheses. Our approach is based on a highly parallelized implementation of the reaction-diffusion equation, and the Gaussian Process Hamiltonian Monte Carlo (GPHMC), a high acceptance rate Monte Carlo technique. We demonstrate our method on synthetic data, and four glioblastoma patients. This promising approach shows that the infiltration is better captured by the model compared to the speed of growth.

Brain Transfer: Spectral Analysis of Cortical Surfaces and Functional Maps.

Abstract: The study of brain functions using fMRI often requires an accurate alignment of cortical data across a population. Particular challenges are surface inflation for cortical visualizations and measurements, and surface matching or alignment of functional data on surfaces for group-level analyses. Present methods typically treat each step separately and can be computationally expensive. For instance, smoothing and matching of cortices often require several hours. Conventional methods also rely on anatomical features to drive the alignment of functional data between cortices, whereas anatomy and function can vary across individuals. To address these issues, we propose BrainTransfer, a spectral framework that unifies cortical smoothing, point matching with confidence regions, and transfer of functional maps, all within minutes of computation. Spectral methods decompose shapes into intrinsic geometrical harmonics, but suffer from the inherent instability of eigenbasis. This limits their accuracy when matching eigenbasis, and prevents the spectral transfer of functions. Our contributions consist of, first, the optimization of a spectral transformation matrix, which combines both, point correspondence and change of eigenbasis, and second, focused harmonics, which localize the spectral decomposition of functional data. BrainTransfer enables the transfer of surface functions across interchangeable cortical spaces, accounts for localized confidence, and gives a new way to perform statistics directly on surfaces. Benefits of spectral transfers are illustrated with a variability study on shape and functional data. Matching accuracy on retinotopy is increased over conventional methods.

Velocity-based cardiac contractility personalization from images using derivative-free optimization

Abstract: Model personalization is a key aspect for biophysical models to impact clinical practice, and cardiac contractility personalization from medical images is a major step in this direction. Existing gradient-based optimization approaches show promising results of identifying the maximum contractility from images, but the contraction and relaxation rates are not accounted for. A main reason is the limited choices of objective functions when their gradients are required. For complicated cardiac models, analytical evalua-tions of gradients are very difficult if not impossible, and finite difference approximations are computationally expensive and may introduce numerical difficulties. By removing such limitations with derivative-free optimization, we found that a velocity-based ob-jective function can properly identify regional maximum contraction stresses, contraction rates, and relaxation rates simultaneously with intact model complexity. Experiments on synthetic data show that the parameters are better identified using the velocity-based objective function than its position-based counterpart, and the proposed framework is insensitive to initial parameters with the adopted derivative-free optimization algorithm. Experiments on clinical data show that the framework can provide personalized contractility parameters which are consistent with the underlying physiologies of the patients and healthy volunteers.

Spectral Forests: Learning of Surface Data, Application to Cortical Parcellation

Abstract: This paper presents a new method for classifying surface data via spectral representations of shapes. Our approach benefits classification problems that involve data living on surfaces, such as in cortical parcellation. For instance, current methods for labeling cortical points into surface parcels often involve a slow mesh deformation toward pre-labeled atlases, requiring as much as 4 hours with the established FreeSurfer. This may burden neuroscience studies involving region-specific measurements. Learning techniques offer an attractive computational advantage, however, their representation of spatial information, typically defined in a Euclidean domain, may be inadequate for cortical parcellation. Indeed, cortical data resides on surfaces that are highly variable in space and shape. Consequently, Euclidean representations of surface data may be inconsistent across individuals. We propose to fundamentally change the spatial representation of surface data, by exploiting spectral coordinates derived from the Laplacian eigenfunctions of shapes. They have the advantage over Euclidean coordinates, to be geometry aware and to parameterize surfaces explicitly. This change of paradigm, from Euclidean to spectral representations, enables a classifier to be applied directly on surface data via spectral coordinates. In this paper, we decide to build upon the successful Random Decision Forests algorithm and improve its spatial representation with spectral features. Our method, Spectral Forests, is shown to significantly improve the accuracy of cortical parcellations over standard Random Decision Forests 74% versus 28% Dice overlaps, and produce accuracy equivalent to FreeSurfer in a fraction of its time 23 seconds versus 3 to 4 hours.

GPSSI: Gaussian Process for Sampling Segmentations of Images

Abstract: Medical image segmentation is often a prerequisite for clinical applications. As an ill-posed problem, it leads to uncertain estimations of the region of interest which may have a significant impact on downstream applications, such as therapy planning. To quantify the uncertainty related to image segmentations, a classical approach is to measure the effect of using various plausible segmentations. In this paper, a method for producing such image segmentation samples from a single expert segmentation is introduced. A probability distribution of image segmentation boundaries is defined as a Gaussian process, which leads to segmentations that are spatially coherent and consistent with the presence of salient borders in the image. The proposed approach outperforms previous generative segmentation approaches, and segmentation samples can be generated efficiently. The sample variability is governed by a parameter which is correlated with a simple DICE score. We show how this approach can have multiple useful applications in the field of uncertainty quantification, and an illustration is provided in radiotherapy planning.

Handbook of biomedical imaging : methodologies and clinical research

Abstract: Object Segmentation and Markov Random Fields.- Fuzzy methods in medical imaging.- Curve Propagation, Level Set Methods and Grouping.- Kernel Methods in Medical Imaging.- Geometric Deformable Models: Overview and Recent Developments.- Active Shape and Appearance Models.- Statistical Atlases.- Statistical Computing on Non-Linear Spaces for Computational Anatomy.- Building Patient-Specific Physical and Physiological Computational Models from Medical Images.- Constructing a Patient-Specific Model Heart from CT Data.- Image-based haemodynamics simulation in intracranial aneursyms.- Atlas-based Segmentation.- Integration of Topological Constraints in Medical Image Segmentation.- Monte Carlo Sampling for the Segmentation of Tubular Structures.- Non-rigid registration using free-form deformations.- Image registration using mutual information.- Physical Model Based Recovery of Displacement and Deformations from 3D medical images.- Cardiovascular Informatics.- Rheumatoid Arthritis Quantifiction using Appearance Models.- Medical Image Processing for Analysis of Colon Motility.- Segmentation of Diseased Livers: A 3D Refinement Approach.- Intra and inter subject analyses of brain functional Magnetic Resonance Images (fMRI).- Diffusion Tensor Estimation, Regularization and Classification.- From Local Q-Ball Estimation to Fibre Crossing Tractography.- Segmentation of clustered cells in microscopy images by geometric PDEs and level sets.- Atlas-based whole-body registration in mice.

Myocardial Infarct Localization Using Neighbourhood Approximation Forests

Abstract: This paper presents a machine-learning algorithm for the automatic localization of myocardial infarct in the left ventricle. Our method constructs neighbourhood approximation forests, which are trained with previously diagnosed 4D cardiac sequences. We introduce a new set of features that simultaneously exploit information from the shape and motion of the myocardial wall along the cardiac cycle. More precisely, characteristics are extracted from a hyper surface that represents the profile of the myocardial thickness. The method has been tested on a database of 65 cardiac MRI images in order to retrieve the diagnosed infarct area. The results demonstrate the effectiveness of the NAF in predicting the left ventricular infarct location in 7 distinct regions. We evaluated our method by verifying the database ground truth. Following a new examination of the 4D cardiac images, our algorithm may detect misclassified infarct locations in the database.

Motion-Aware Mosaicing for Confocal Laser Endomicroscopy

Abstract: Probe-based Confocal Laser Endomicroscopy pCLE provides physicians with real-time access to histological information during standard endoscopy procedures, through high-resolution cellular imaging of internal tissues. Earlier work on mosaicing has enhanced the potential of this imaging modality by meeting the need to get a complete representation of the imaged region. However, with approaches, the dynamic information, which may be of clinical interest, is lost. In this study, we propose a new mosaic construction algorithm for pCLE sequences based on a min-cut optimization and gradient-domain composition. Its main advantage is that the motion of some structures within the tissue such as blood cells in capillaries, is taken into account. This allows physicians to get both a sharper static representation and a dynamic representation of the imaged tissue. Results on 16 sequences acquired in vivo on six different organs demonstrate the clinical relevance of our approach.

Propagation of Myocardial Fibre Architecture Uncertainty on Electromechanical Model Parameter Estimation: A Case Study

Abstract: Computer models of the heart are of increasing interest for clinical applications due to their discriminative and predictive power. However the personalisation step to go from a generic model to a patient-specific one is still a scientific challenge. In particular it is still difficult to quantify the uncertainty on the estimated parameters and predicted values. In this manuscript we present a new pipeline to evaluate the impact of fibre uncertainty on the personalisation of an electromechanical model of the heart from ECG and medical images. We detail how we estimated the variability of the fibre architecture among a given population and how the uncertainty generated by this variability impacts the following personalisation. We first show the variability of the personalised simulations, with respect to the principal variations of the fibres. Then discussed how the variations in this (small) healthy population of fibres impact the parameters of the personalised simulations.

Expert-validated CSF segmentation of MNI atlas enhances accuracy of virtual glioma growth patterns

Abstract: Biomathematical modeling of glioma growth has been developed to optimize treatments delivery and to evaluate their efficacy. Simulations currently make use of anatomical knowledge from standard MRI atlases. For example, cerebrospinal fluid (CSF) spaces are obtained by automatic thresholding of the MNI atlas, leading to an approximate representation of real anatomy. To correct such inaccuracies, an expert-revised CSF segmentation map of the MNI atlas was built. Several virtual glioma growth patterns of different locations were generated, with and without using the expert-revised version of the MNI atlas. The adequacy between virtual and radiologically observed growth patterns was clearly higher when simulations were based on the expert-revised atlas. This work emphasizes the need for close collaboration between clinicians and researchers in the field of brain tumor modeling.

Joint Morphometry of Fiber Tracts and Gray Matter Structures Using Double Diffeomorphisms.

Abstract: This work proposes an atlas construction method to jointly analyse the relative position and shape of fiber tracts and gray matter structures. It is based on a double diffeomorphism which is a composition of two diffeomorphisms. The first diffeomorphism acts only on the white matter keeping fixed the gray matter of the atlas. The resulting white matter , together with the gray matter , are then deformed by the second diffeomorphism. The two diffeomorphisms are related and jointly optimised. In this way , the first diffeomorphisms explain the variability in structural connectivity within the population , namely both changes in the connected areas of the gray matter and in the geometry of the pathway of the tracts. The second diffeomorphisms put into correspondence the homologous anatomical structures across subjects. Fiber bundles are approximated with weighted prototypes using the metric of weighted currents . The atlas , the covariance matrix of deformation parameters and the noise variance of each structure are automatically estimated using a Bayesian approach. This method is applied to patients with Tourette syndrome and controls showing a variability in the structural connectiv-ity of the left cortico-putamen circuit .

The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)

Abstract: In this paper we report the set-up and results of the Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) organized in conjunction with the MICCAI 2012 and 2013 conferences. Twenty state-of-the-art tumor segmentation algorithms were applied to a set of 65 multi-contrast MR scans of low- and high-grade glioma patients—manually annotated by up to four raters—and to 65 comparable scans generated using tumor image simulation software. Quantitative evaluations revealed considerable disagreement between the human raters in segmenting various tumor sub-regions (Dice scores in the range 74%–85%), illustrating the difficulty of this task. We found that different algorithms worked best for different sub-regions (reaching performance comparable to human inter-rater variability), but that no single algorithm ranked in the top for all sub-regions simultaneously. Fusing several good algorithms using a hierarchical majority vote yielded segmentations that consistently ranked above all individual algorithms, indicating remaining opportunities for further methodological improvements. The BRATS image data and manual annotations continue to be publicly available through an online evaluation system as an ongoing benchmarking resource.

Building Patient-Specific Physical and Physiological Computational Models from Medical Images

Abstract: We describe a hierarchy of computational models of the human body operating at the geometrical, physical and physiological levels. Those models can be coupled with medical images which play a crucial role in the diagnosis, planning, control and follow-up of therapy. In this paper, we discuss the issue of building patient-specific physical and physiological models from macroscopic observations extracted from medical images. We illustrate the topic of model personalization with concrete examples in brain shift modeling, hepatic surgery simulation, cardiac and tumor growth modeling. We conclude this article with scientific perspectives.

Local late gadolinium enhancement features to identify the electrophysiological substrate of post-infarction ventricular tachycardia: a machine learning approach

Abstract: Background Most ventricular tachycardias occur on structurally diseased hearts with fibrotic scar, where bundles of surviving tissue promote electrical circuit re-entry. These bundles can be identified on invasive electrophysiological (EP) mapping as local abnormal ventricular activities (LAVA) during sinus rhythm. Although the elimination of LAVAs by radiofrequency ablation was shown to be an efficient therapeutic option, their identification requires is a lengthy and invasive process. Late gadolinium enhancement (LGE) magnetic resonance imaging enables a non-invasive 3D assessment of scar topology and heterogeneity with millimetric spatial resolution. The aim of this work is to identify imaging features associated with LAVA, features that may subsequently be used to target ablation or to stratify the risk of arrhythmia.

Unified analysis of shape and structural connectivity of neural pathways

Abstract: An abnormal brain development due to a neuropsychiatric disorder can influence the shape and the anatomical organization of both white and grey matter structures. An example is the syndrome of Gilles de la Tourette (GTS) which is thought to be associated with dysfunctions of the cortico-striato-pallido-thalamic circuits [6]. These anatomical complexes should be studied as a whole, analysing both the shape and the relative position of their structures. Atlas constructions permit to estimate an average shape complex of a given population, called template, and its deformations towards the shape complexes of each subject. The template represents the morphological invariants of the population whereas the deformations capture its variability. Previous works defined these deformations as single diffeomorphisms acting on the entire 3D space, so that ending points of fiber bundles could not move independently of grey matter structures [1,2,4,5]. This implicitly assumes that fiber bundles connect the same areas of grey matter structures across subjects. This assumption is not compatible with the aforementioned hypothesis about GTS [6] which relates the syndrome to atypical configurations of neural circuits. We propose a new atlas construction method which can handle both fibers and surfaces and which is based on a double diffeomorphism. This permits to analyse the morphological variations of each structure and the changes in the relative position between fiber bundles and grey matter structures, namely the variations in structural connectivity.