Catálogo de publicaciones - libros

Preoperative planning systems are commonly used for oral implant surgery. One of the objectives is to determine if the quantity and quality of bone is sufficient to sustain an implant while avoiding critical anatomic structures. We aim to automate the segmentation of jaw tissues on CT images: cortical bone, trabecular core and especially the mandibular canal containing the dental nerve. This nerve must be avoided during implant surgery to prevent lip numbness. Previous work in this field used thresholds or filters and needed manual initialization. An automated system based on the use of Active Appearance Models (AAMs) is proposed. Our contribution is a completely automated segmentation of tissues and a semi-automatic landmarking process necessary to create the AAM model. The AAM is trained using 215 images and tested with a leave-4-out scheme. Results obtained show an initialization error of 3.25% and a mean error of 1.63mm for the cortical bone, 2.90mm for the trabecular core, 4.76mm for the mandibular canal and 3.40mm for the dental nerve.

In this paper, we present a comprehensive framework to detect morphological changes in skull vaults of adolescent idiopathic scoliosis girls. To our knowledge, this is the first attempt to use a combination of medical knowledge, image analysis techniques, statistical learning tools, and scientific visualization methods to detect skull morphological changes. The shape analysis starts from a reliable 3-D segmentation of the skull using thresholding and math-morphological operations. The gradient vector flow is used to model the skull vault surface, which is followed by a spherically uniform sampling. The scale-normalized distances from the shape centroid to sample points are defined as the features. The most discriminative features are selected using recursive feature elimination for support vector machine. The results of this study specify the skull vault surface changes and shed light on building the evidence of bone formation abnormality in AIS girls.

Cortical surface reconstruction is important for functional brain mapping and morphometric analysis of the brain cortex. Several methods have been developed for the faithful reconstruction of surface models which describe the true cortical surface in both geometry and topology. However there has been no explicit method for the quantitative evaluation of the whole-cortical-surface models. In this study, we present a novel phantom-based evaluation method of the cortical surface reconstruction algorithm and quantitatively validated the local morphometric accuracy of CLASP which is one of the well-established reconstruction methods. The evaluation included local geometrical accuracy and performance of cortical thickness measure. The validation study revealed that there were some underestimations of cortical thickness measure using CLASP in the ventral and sulcal areas of the cortex and overestimations in the gyral areas and inferior temporal lobe. This study could present a generic metric for the quantitative evaluation of cortical surface reconstruction algorithm.

Tensor-based morphometry (TBM) is widely used in computational anatomy as a means to understand shape variation between structural brain images. A 3D nonlinear registration technique is typically used to align all brain images to a common neuroanatomical template, and the deformation fields are analyzed statistically to identify group differences in anatomy. However, the differences are usually computed solely from the determinants of the Jacobian matrices that are associated with the deformation fields computed by the registration procedure. Thus, much of the information contained within those matrices gets thrown out in the process. Only the magnitude of the expansions or contractions is examined, while the anisotropy and directional components of the changes are ignored.

Here we remedy this problem by computing multivariate shape change statistics using the strain matrices. As the latter do not form a vector space, means and covariances are computed on the manifold of positive-definite matrices to which they belong. We study the brain morphology of 26 HIV/AIDS patients and 14 matched healthy control subjects using our method.

The images are registered using a high-dimensional 3D fluid registration algorithm, which optimizes the Jensen-Rényi divergence, an information-theoretic measure of image correspondence. The anisotropy of the deformation is then computed. We apply a manifold version of Hotelling’s test to the strain matrices. Our results complement those found from the determinants of the Jacobians alone and provide greater power in detecting group differences in brain structure.

The segmentation of newborn brain MRI is important for assessing and directing treatment options for premature infants at risk for developmental disorders, abnormalities, or even death. Segmentation of infant brain MRI is particularly challenging when compared with the segmentation of images acquired from older children and adults. We sought to develop a fully automated segmentation strategy and present here a Bayesian approach utilizing an atlas of priors derived from previous segmentations and a new scheme for automatically selecting and iteratively refining classifier training data using the STAPLE algorithm. Results have been validated by comparison to hand-drawn segmentations.

This work investigates the effects of nonrigid transformation model and deformation constraints on the results of deformation-based morphometry (DBM) studies. We evaluate three popular registration algorithms: a B-spline algorithm with several different constraint terms, Thirion’s demons algorithm, and a curvature PDE-based algorithm. All algorithms produced virtually identical overlaps of corresponding structures, but the underlying deformation fields were very different, and the Jacobian determinant values within homogeneous structures varied dramatically. In several cases, we observed bi-modal distributions of Jacobians within a region that violate the assumption of gaussianity that underlies many statistical tests. Our results demonstrate that, even with perfect overlap of corresponding structures, the statistics of Jacobian values are affected by bias due to design elements of the particular nonrigid registration. These findings are not limited to DBM, but also apply to voxel-based morphometry to the extent that it includes a Jacobian-based correction step (“modulation”).

This work studies limits on estimating the width of tubular structures in 3D images. Based on nonlinear estimation theory we analyze the minimal stochastic error of estimating the width. Given a 3D analytic model of the image intensities of tubular structures, we derive a closed-form expression for the Cramér-Rao bound of the width estimate under image noise. We use the derived lower bound as a benchmark and compare it with three previously proposed accuracy limits for vessel width estimation. Moreover, by experimental investigations we demonstrate that the derived lower bound can be achieved by fitting a 3D parametric intensity model directly to the image data.

We propose a novel trace-based method to extract vessel axes from gray-scale angiograms without preliminary segmentations. Our method traces the axes on an optimization framework with the bounded spherical projection images and the sum of squared difference metric. It does not take alternate steps to search the next axial point and its tangent as in other trace-based algorithms, instead the novel method finds the solution simultaneously. This helps avoid U-turns of the trace and large spatial discontinuity of the axial points. Another advantage of the method is that it enables interactive user guidance to produce continuous tracing through regions that contain furcations, disease portions, kissing vessels (vessels in close proximity to each other) and thin vessels, which pose difficulties for the other algorithms and make re-initialization inevitable as illustrated on synthetic and clinical data sets.

In this article, we present a general statistical parts-based model for representing the appearance of an image set, applied to the problem of inter-subject MR brain image matching. In contrast with global image representations such as active appearance models, the parts-based model consists of a collection of localized image parts whose appearance, geometry and occurrence frequency are quantified statistically. The parts-based approach explicitly addresses the case where one-to-one correspondence does not exist between subjects due to anatomical differences, as parts are not expected to occur in all subjects. The model can be learned automatically, discovering structures that appear with statistical regularity in a large set of subject images, and can be robustly fit to new images, all in the presence of significant inter-subject variability. As parts are derived from generic scale-invariant features, the framework can be applied in a wide variety of image contexts, in order to study the commonality of anatomical parts or to group subjects according to the parts they share. Experimentation shows that a parts-based model can be learned from a large set of MR brain images, and used to determine parts that are common within the group of subjects. Preliminary results indicate that the model can be used to automatically identify distinctive features for inter-subject image registration despite large changes in appearance.

The support vector domain description is a one-class classification method that estimates the shape and extent of the distribution of a data set. This separates the data into outliers, outside the decision boundary, and inliers on the inside. The method bears close resemblance to the two-class support vector machine classifier. Recently, it was shown that the regularization path of the support vector machine is piecewise linear, and that the entire path can be computed efficiently. This paper shows that this property carries over to the support vector domain description. Using our results the solution to the one-class classification can be obtained for any amount of regularization with roughly the same computational complexity required to solve for a particularly value of the regularization parameter. The possibility of evaluating the results for any amount of regularization not only offers more accurate and reliable models, but also makes way for new applications. We illustrate the potential of the method by determining the order of inclusion in the model for a set of corpora callosa outlines.