Results 1  10
of
30
Joint Segmentation of Image Ensembles via Latent Atlases
, 2009
"... Spatial priors, such as probabilistic atlases, play an important role in MRI segmentation. However, the availability of comprehensive, reliable and suitable manual segmentations for atlas construction is limited. We therefore propose a joint segmentation of corresponding, aligned structures in the ..."
Abstract

Cited by 23 (9 self)
 Add to MetaCart
(Show Context)
Spatial priors, such as probabilistic atlases, play an important role in MRI segmentation. However, the availability of comprehensive, reliable and suitable manual segmentations for atlas construction is limited. We therefore propose a joint segmentation of corresponding, aligned structures in the entire population that does not require a probability atlas. Instead, a latent atlas, initialized by a single manual segmentation, is inferred from the evolving segmentations of the ensemble. The proposed method is based on probabilistic principles but is solved using partial differential equations (PDEs) and energy minimization criteria. We evaluate the method by segmenting 50 brain MR volumes. Segmentation accuracy for cortical and subcortical structures approaches the quality of stateoftheart atlasbased segmentation results, suggesting that the latent atlas method is a reasonable alternative when existing atlases are not compatible with the data to be processed.
Convex MultiRegion Probabilistic Segmentation with Shape Prior in the Isometric LogRatio Transformation Space
"... Image segmentation is often performed via the minimization of an energy function over a domain of possible segmentations. The effectiveness and applicability of such methods depends greatly on the properties of the energy function and its domain, and on what information can be encoded by it. Here we ..."
Abstract

Cited by 10 (6 self)
 Add to MetaCart
(Show Context)
Image segmentation is often performed via the minimization of an energy function over a domain of possible segmentations. The effectiveness and applicability of such methods depends greatly on the properties of the energy function and its domain, and on what information can be encoded by it. Here we propose an energy function that achieves several important goals. Specifically, our energy function is convex and incorporates shape prior information while simultaneously generating a probabilistic segmentation for multiple regions. Our energy function represents multiregion probabilistic segmentations as elements of a vector space using the isometric logratio (ILR) transformation. To our knowledge, these four goals (convex, with shape priors, multiregion, and probabilistic) do not exist together in any other method, and this is the first time ILR is used in an image segmentation method. We provide examples demonstrating the usefulness of these features. 1.
Medialbased Deformable Models in Nonconvex Shapespaces for Medical Image Segmentation using Genetic Algorithms
"... We explore the application of genetic algorithms (GA) to deformable models through the proposition of a novel method for medical image segmentation that combines GA with nonconvex, localized, medialbased shape statistics. We replace the more typical gradient descent optimizer used in deformable m ..."
Abstract

Cited by 10 (3 self)
 Add to MetaCart
(Show Context)
We explore the application of genetic algorithms (GA) to deformable models through the proposition of a novel method for medical image segmentation that combines GA with nonconvex, localized, medialbased shape statistics. We replace the more typical gradient descent optimizer used in deformable models with GA, and the convex, implicit, global shape statistics with nonconvex, explicit, localized ones. Specifically, we propose GA to reduce typical deformable model weaknesses pertaining to model initialization, pose estimation and local minima, through the simultaneous evolution of a large number of models. Furthermore, we constrain the evolution, and thus reduce the size of the searchspace, by using statisticallybased deformable models whose deformations are intuitive (stretch, bulge, bend) and are driven in terms of localized principal modes of variation, instead of modes of variation across the entire shape that often fail to capture localized shape changes. Although GA are not guaranteed to achieve the global optima, our method compares favorably to the prevalent optimization techniques, convex / nonconvex gradientbased optimizers and to globally optimal graphtheoretic combinatorial optimization techniques, when applied to the task of corpus callosum segmentation in 50 midsagittal brain magnetic resonance images.
Probabilistic multishape segmentation of knee extensor and flexor muscles
 In MICCAI
, 2011
"... Abstract. Patients with chronic obstructive pulmonary disease (COPD) often exhibit skeletal muscle weakness in lower limbs. Analysis of the shapes and sizes of these muscles can lead to more effective therapy. Unfortunately, segmenting these muscles from one another is a challenging task due to a la ..."
Abstract

Cited by 8 (5 self)
 Add to MetaCart
(Show Context)
Abstract. Patients with chronic obstructive pulmonary disease (COPD) often exhibit skeletal muscle weakness in lower limbs. Analysis of the shapes and sizes of these muscles can lead to more effective therapy. Unfortunately, segmenting these muscles from one another is a challenging task due to a lack of image information in many areas. We present a fully automatic segmentation method that overcomes the inherent difficulties of this problem to accurately segment the different muscles. Our method enforces a multiregion shape prior on the segmentation to ensure feasibility and provides an energy minimizing probabilistic segmentation that indicates areas of uncertainty. Our experiments on 3D MRI datasets yield an average Dice similarity coefficient of 0.92 ± 0.03 with the ground truth. 1
Probabilistic MultiShape Representation using an Isometric LogRatio Mapping
"... Abstract. Several sources of uncertainties in shape boundaries in medical images have motivated the use of probabilistic labeling approaches. Although it is wellknown that the sample space for the probabilistic representation of a pixel is the unit simplex, standard techniques of statistical shape ..."
Abstract

Cited by 7 (7 self)
 Add to MetaCart
(Show Context)
Abstract. Several sources of uncertainties in shape boundaries in medical images have motivated the use of probabilistic labeling approaches. Although it is wellknown that the sample space for the probabilistic representation of a pixel is the unit simplex, standard techniques of statistical shape analysis (e.g. principal component analysis) have been applied to probabilistic data as if they lie in the unconstrained real Euclidean space. Since these techniques are not constrained to the geometry of the simplex, the statistically feasible data produced end up representing invalid (out of the simplex) shapes. By making use of methods for dealing with what is known as compositional or closed data, we propose a new framework intrinsic to the unit simplex for statistical analysis of probabilistic multishape anatomy. In this framework, the isometric logratio (ILR) transformation is used to isometrically and bijectively map the simplex to the Euclidean real space, where data are analyzed in the same way as unconstrained data and then backtransformed to the simplex. We demonstrate favorable properties of ILR over existing mappings (e.g. LogOdds). Our results on synthetic and brain data exhibit a more accurate statistical analysis of probabilistic shapes. 1
Label Space  A Coupled MultiShape Representation
"... Richly labeled images representing several substructures of an organ occur quite frequently in medical images. For example, a typical brain image can be labeled into grey matter, white matter or cerebrospinal fluid, each of which may be subdivided further. Many manipulations such as interpolation, ..."
Abstract

Cited by 6 (0 self)
 Add to MetaCart
Richly labeled images representing several substructures of an organ occur quite frequently in medical images. For example, a typical brain image can be labeled into grey matter, white matter or cerebrospinal fluid, each of which may be subdivided further. Many manipulations such as interpolation, transformation, smoothing, or registration need to be performed on these images before they can be used in further analysis. In this work, we present a novel multishape representation and compare it with the existing representations to demonstrate certain advantages of using the proposed scheme. Specifically, we propose label space, a representation that is both flexible and well suited for coupled multishape analysis. Under this framework, object labels are mapped to vertices of a regular simplex, e.g. the unit interval for two labels, a triangle for three labels, a tetrahedron for four labels, etc. This forms the basis of a convex linear structure with the property that all labels are equally spaced. We will demonstrate that this representation has several desirable properties: algebraic operations may be performed directly, label uncertainty is expressed equivalently as a weighted mixture of labels or in a probabilistic manner, and interpolation is unbiased toward any label or the background. In order to demonstrate these properties, we compare label space to signed distance maps as well as other implicit representations in tasks such as smoothing, interpolation, registration, and principal component analysis.
Label space: A multiobject shape representation. Combinatorial Image Analysis
, 2008
"... Two key aspects of coupled multiobject shape analysis are the choice of representation and subsequent registration to align the sample set. Current techniques for such analysis tend to trade off performance between the two tasks, performing well for one task but developing problems when used for t ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
(Show Context)
Two key aspects of coupled multiobject shape analysis are the choice of representation and subsequent registration to align the sample set. Current techniques for such analysis tend to trade off performance between the two tasks, performing well for one task but developing problems when used for the other. This article proposes L n label space, a representation that is both flexible and well suited for both tasks. We propose to map object labels to vertices of a regular simplex, e.g. the unit interval for two labels, a triangle for three labels, a tetrahedron for four labels, etc. This forms a linear space with the property that all labels are equally separated. On examination, this representation has several desirable properties: algebraic operations may be done directly, label uncertainty is expressed as a weighted mixture of labels, interpolation is unbiased toward any label or the background, and registration may be performed directly. To demonstrate these properties, we describe variational registration directly in this space. Many registration methods fix one of the maps and align the rest of the set to this fixed map. To remove the bias induced by arbitrary selection of the fixed map, we align a set of label maps to their intrinsic mean map.
Markov Dependence TreeBased Segmentation of Deep Brain Structures
"... Abstract. We propose a new framework for multiobject segmentation of deep brain structures, which have significant shape variations and relatively small sizes in medical brain images. In the images, the structure boundaries may be blurry or even missing, and the surrounding background is a clutter ..."
Abstract

Cited by 3 (1 self)
 Add to MetaCart
(Show Context)
Abstract. We propose a new framework for multiobject segmentation of deep brain structures, which have significant shape variations and relatively small sizes in medical brain images. In the images, the structure boundaries may be blurry or even missing, and the surrounding background is a clutter and full of irrelevant edges. We suggest a templatebased framework, which fuses the information of edge features, region statistics and interstructure constraints to detect and locate all the targeted brain structures such that manual initialization is unnecessary. The multiobject template is organized in the form of a hierarchical Markov dependence tree. It makes the matching of multiple objects efficient. Our approach needs only one example as training data and alleviates the demand of a large training set. The obtained segmentation results on real data are encouraging and the proposed method enjoys several important advantages over existing methods. 1
Joint Segmentation via PatientSpecific Latent Anatomy Model
 PROBABILISTIC MODELS FOR MEDICAL IMAGE ANALYSIS
, 2009
"... We present a generative approach for joint 3D segmentation of patientspecific MR scans across different modalities or time points. The latent anatomy, in the form of spatial parameters, is inferred simultaneously with the evolution of the segmentations. The individual segmentation of each scan sup ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
(Show Context)
We present a generative approach for joint 3D segmentation of patientspecific MR scans across different modalities or time points. The latent anatomy, in the form of spatial parameters, is inferred simultaneously with the evolution of the segmentations. The individual segmentation of each scan supports the segmentation of the group by sharing common information. The joint segmentation problem is solved via a statistically driven levelset framework. We illustrate the method on an example application of multimodal and longitudinal brain tumor segmentation, reporting promising segmentation results.
Characterization of Anatomical Shape Based on Random Walk Hitting Times
"... Abstract. This paper presents an implicit shape representation for describing anatomical shapes with high interpatient variability based on the expected boundary hitting time of a random walk, which happens to be the solution to the Poisson equation. The main contribution of this paper is to test t ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
(Show Context)
Abstract. This paper presents an implicit shape representation for describing anatomical shapes with high interpatient variability based on the expected boundary hitting time of a random walk, which happens to be the solution to the Poisson equation. The main contribution of this paper is to test the validity of the Poissonbased mapping for representing anatomical shape, comparing its compactness and completeness with the commonly used Signed Distance Transform and using the liver and the caudate nucleus as examples. Based on these findings, we discuss its use as a shape prior for image segmentation. 1