Abstract not found

Statistical analysis of anatomical maps in a stereotaxic space has been shown to be a useful tool in population-based studies for quantifying local anatomical differences or changes, without a priori assumptions about the location and extent of the regions of interest. This paper presents an extension and validation of a previously published methodology, referred to as RAVENS, for characterizing regional atrophy in the brain. A new method for elastic, volume-preserving spatial normalization, which allows for accurate quantification of very localized atrophy, is used. The RAVENS methodology was tested on images with simulated atrophy within two gyri: precentral and superior temporal. It was found to accurately determine the regions of atrophy, despite their localized nature and the interindividual variability of cortical structures. Moreover, it was found to perform substantially better than the voxel-based morphology method of SPM’99. Improved sensitivity was achieved at the expense of human effort involved in defining a number of sulcal curves that serve as constraints on the 3D elastic warping. © 2001 Academic Press

Due to the finite spatial resolution of imaging devices, a single voxel in a medical image may be composed of mixture of tissue types, an effect known as partial volume effect (PVE). Partial volume estimation, that is, the estimation of the amount of each tissue type within each voxel, has received considerable interest in recent years. Much of this work has been focused on the mixel model, a statistical model of PVE. We propose a novel trimmed minimum covariance determinant (TMCD) method for the estimation of the parameters of the mixel PVE model. In this method, each voxel is first labeled according to the most dominant tissue type. Voxels that are prone to PVE are removed from this labeled set, following which robust location estimators with high breakdown points are used to estimate the mean and the covariance of each tissue class. Comparisons between different methods for parameter estimation based on classified images as well as expectation–maximization-like (EM-like) procedure for simultaneous parameter and

We propose a model of appearance and a matching method which combines `global' models (in which a few parameters control global appearance) with local elastic or optical-flow-based methods, in which deformation is described by many local parameters together with some regularisation constraints. We use an Active Appearance Model (AAM) as the global model, which can match a statistical model of appearance to a new image rapidly. However, the amount of variation allowed is constrained by the modes of the model, which may be too restrictive (for instance when insufficient training examples are available, or the number of modes is deliberately truncated for efficiency or memory conservation). To compensate for this, after global AAM convergence, we allow further local model deformation, driven by local AAMs around each model node. This is analogous to optical ow or `demon' methods of non-linear image registration. We describe the technique in detail, and demonstrate that allowing...

Abstract. We introduce a novel approach for magnetic resonance image (MRI) brain tissue classification by learning image neighborhood statistics from noisy input data using nonparametric density estimation. The method models images as random fields and relies on minimizing an entropy-based metric defined on high dimensional probability density functions. Combined with an atlas-based initialization, it is completely automatic. Experiments on real and simulated data demonstrate the advantages of the method in comparison to other approaches. 1

Abstract. This paper describes an hybrid method combining symbolic and numerical techniques for annotating brain Magnetic Resonance images. Existing automatic labelling methods are mostly statistical in nature and do not work very well in certain situations such as the presence of lesions. The goal is to assist them by a knowledge-based method. The system uses statistical method for generating a sufficient set of initial facts for fruitful reasoning. Then, the reasoning is supported by an OWL DL ontology enriched by SWRL rules. The experiments described were achieved using the KAON2 reasoner for inferring the annotations. 1

Abstract. The concept of the Logarithm of the Odds (LogOdds) is frequently used in areas such as artificial neural networks, economics, and biology. Here, we utilize LogOdds for a shape representation that demonstrates desirable properties for medical imaging. For example, the representation encodes the shape of an anatomical structure as well as the variations within that structure. These variations are embedded in a vector space that relates to a probabilistic model. We apply our representation to a voxel based segmentation algorithm. We do so by embedding the manifold of Signed Distance Maps (SDM) into the linear space of LogOdds. The LogOdds variant is superior to the SDM model in an experiment segmenting 20 subjects into subcortical structures. We also use LogOdds in the non-convex interpolation between space conditioned distributions. We apply this model to a longitudinal schizophrenia study using quadratic splines. The resulting time-continuous simulation of the schizophrenic aging process has a higher accuracy then a model based on convex interpolation. 1

Automatic segmentation of MR images of the developing

Standard image based segmentation approaches perform poorly when there is little or no contrast along boundaries of different regions. In such cases, segmentation is largely performed manually using prior knowledge of the shape and relative location of the underlying structures combined with partially discernible boundaries. We present an automated approach guided by covariant shape deformations of neighboring structures, which is an additional source of prior information. Captured by a shape atlas, these deformations are transformed into a statistical model using the logistic function. Structure boundaries, anatomical labels, and image inhomogeneities are estimated simultaneously within an Expectation-Maximization formulation of the maximum a posteriori probability estimation problem. We demonstrate the approach on 20 brain magnetic resonance images showing superior performance, particularly in cases where purely image based methods fail.

Labelling brain images content at the semantic level is important for decision support in the context of neuroimaging and neurosurgery, as well as for providing images annotations that may support future retrieval. This paper shows how symbolic methods can be used for the semantic description of the images, and the interest of combining ontologies and rules for it. A simplified example illustrates the method proposed for assisting the labelling of some brain structures in Magnetic Resonance Imaging images. 1