Catálogo de publicaciones - libros

Energy functional minimization is an increasingly popular technique for image segmentation. However, it is far too commonly applied with hand-tuned parameters and initializations that have only been validated for a few images. Fixing these parameters over a set of images assumes the same parameters are ideal for each image. We highlight the effects of varying the parameters and initialization on segmentation accuracy and propose a framework for attaining improved results using image adaptive parameters and initializations. We provide an analytical definition of optimal weights for functional terms through an examination of segmentation in the context of image manifolds, where nearby images on the manifold require similar parameters and similar initializations. Our results validate that fixed parameters are insufficient in addressing the variability in real clinical data, that similar images require similar parameters, and demonstrate how these parameters correlate with the image manifold. We present significantly improved segmentations for synthetic images and a set of 470 clinical examples.

Nonlinear image registration is a challenging task in the field of medical image analysis. In many applications discontinuities may be present in the displacement field, and intensity variations may occur. In this work we therefore utilize an energy functional which is based on Total Variation regularization and a robust data term. We propose a novel, fast and stable numerical scheme to find the minimizer of this energy. Our approach combines a fixed-point procedure derived from duality principles combined with a fast thresholding step. We show experimental results on synthetic and clinical CT lung data sets at different breathing states as well as registration results on inter-subject brain MRIs.

Our paper summarizes experiments for measuring the accuracy of deformable 2D-3D registration between sets of simulated x-ray images (DRR’s) and a statistical shape model of the pelvis bones, which includes x-ray attenuation information (“density”). In many surgical scenarios, the images contain a truncated view of the pelvis anatomy. Our work specifically addresses this problem by examining different selections of truncated views as target images. Our atlas is derived by applying principal component analysis to a population of up to 110 instance shapes. The experiments measure the registration error with a large and truncated FOV. A typical accuracy of about 2 mm is achieved in the 2D-3D registration, compared with about 1.4 mm of an “optimal” 3D-3D registration.

2D-3D registration of abdominal angiographic data is a difficult problem due to hard time constraints during the intervention, different vessel contrast in volume and image, and motion blur caused by breathing. We propose a novel method for aligning 2D Digitally Subtracted Angiograms (DSA) to Computed Tomography Angiography (CTA) volumes, which requires no user interaction intrainterventionally. In an iterative process, we link 2D segmentation and 2D-3D registration using a probability map, which creates a common feature space where outliers in 2D and 3D are discarded consequently. Unlike other approaches, we keep user interaction low while high capture range and robustness against vessel variability and deformation are maintained. Tests on five patient data sets and a comparison to two recently proposed methods show the good performance of our method.

In this paper we propose a novel approach for automatic segmentation of cartilage using a statistical atlas and efficient primal/dual linear programming. To this end, a novel statistical atlas construction is considered from registered training examples. Segmentation is then solved through registration which aims at deforming the atlas such that the conditional posterior of the learned (atlas) density is maximized with respect to the image. Such a task is reformulated using a discrete set of deformations and segmentation becomes equivalent to finding the set of local deformations which optimally match the model to the image. We evaluate our method on 56 MRI data sets (28 used for the model and 28 used for evaluation) and obtain a fully automatic segmentation of patella cartilage volume with an overlap ratio of 0.84 with a sensitivity and specificity of 94.06% and 99.92%, respectively.

The use of groupwise registration techniques for average atlas construction has been a growing area of research in recent years. One particularly challenging component of groupwise registration is finding scalable and effective groupwise similarity metrics; these do not always extend easily from pairwise metrics. This paper investigates possible choices of similarity metrics and additionally proposes a novel metric based on Normalised Mutual Information. The described groupwise metrics are quantitatively evaluated on simulated and 3D MR datasets, and their performance compared to equivalent pairwise registration.

During neurosurgery, brain deformation renders preoperative images unreliable for localizing pathologic structures. In order to visualize the current brain anatomy, it is necessary to nonrigidly warp these preoperative images to reflect the intraoperative brain. This can be accomplished using a biomechanical model driven by sparse intraoperative information. In this paper, a linear elastic model of the brain is developed which can infer volumetric brain deformation given the cortical surface displacement. This model was tested on both a realistic brain phantom and , proving its ability to account for large brain deformations. Also, an efficient semiautomatic strategy for preoperative cortical feature detection is outlined, since accurate segmentation of cortical features can aid intraoperative cortical surface tracking.

This paper presents a Bayesian framework for tracking of tubular structures such as vessels. Compared to conventional tracking schemes, its main advantage is its non-deterministic character, which strongly increases the robustness of the method. A key element of our approach is a dedicated observation model for tubular structures in regions with varying intensities. Furthermore, we show how the tracking method can be used to obtain a probabilistic segmentation of the tracked tubular structure. The method has been applied to track the internal carotid artery from CT angiography data of 14 patients (28 carotids) through the skull base. This is a challenging problem, owing to the close proximity of bone, overlap in intensity values of lumen voxels and (partial volume) bone voxels, and the tortuous path of the vessels. The tracking was successful in 25 cases, and the extracted path were found to be close (< 1.0mm) to manually traced paths by two observers.

Automatic delineation and robust measurement of fetal anat-omical structures in 2D ultrasound images is a challenging task due to the complexity of the object appearance, noise, shadows, and quantity of information to be processed. Previous solutions rely on explicit encoding of prior knowledge and formulate the problem as a perceptual grouping task solved through clustering or variational approaches. These methods are known to be limited by the validity of the underlying assumptions and cannot capture complex structure appearances. We propose a novel system for fast automatic obstetric measurements by directly exploiting a large database of expert annotated fetal anatomical structures in ultrasound images. Our method learns to distinguish between the appearance of the object of interest and background by training a discriminative constrained probabilistic boosting tree classifier. This system is able to handle previously problems in this domain, such as the effective segmentation of fetal abdomens. We show results on fully automatic measurement of head circumference, biparietal diameter, abdominal circumference and femur length. Unparalleled extensive experiments show that our system is, on average, close to the accuracy of experts in terms of segmentation and obstetric measurements. Finally, this system runs under half second on a standard dual-core PC computer.

A methodology is introduced for the automated assessment of structural changes of breast tissue in mammograms. It employs a generic machine learning framework and provides objective breast density measures quantifying the specific biological effects of interest. In several illustrative experiments on data from a clinical trial, it is shown that the proposed method can quantify effects caused by hormone replacement therapy (HRT) at least as good as standard methods. Most interestingly, the separation of subpopulations using our approach is considerably better than the best alternative, which is interactive. Moreover, the automated method is capable of detecting age effects where standard methodologies completely fail.