Segmentation of anatomical structures in medical imagery is a key step in a variety of clinical applications. Designing a generic, automated method that works for various structures and imaging modalities is a daunting task. In this paper, we present an effective interactive segmentation method that reformulates the GrowCut algorithm as a clustering problem and computes a fast, approximate solution. The method is further improved by using an efficient updating scheme requiring only local computations when new user input becomes available, making it applicable to high resolution images. The algorithm may easily be included as a user-oriented software module in any number of available medical imaging/image processing platforms such as 3D Slicer. The efficiency and effectiveness of the algorithm are demonstrated through tests on several challenging data sets where it is also compared to standard GrowCut.

Traditional models of the human language circuitry encompass three cortical areas, Broca s, Geschwind s and Wernicke s, and their connectivity through white matter fascicles. The neural connectivity deep to these cortical areas remains poorly understood, as does the macroscopic functional organization of the cortico-subcortical language circuitry. In an effort to expand current knowledge, we combined functional MRI (fMRI) and diffusion tensor imaging to explore subject-specific structural and functional macroscopic connectivity, focusing on Broca s area. Fascicles were studied using diffusion tensor imaging fiber tracking seeded from volumes placed manually within the white matter. White matter fascicles and fMRI-derived clusters (antonym-generation task) of positive and negative blood-oxygen-level-dependent (BOLD) signal were co-registered with 3-D renderings of the brain in 12 healthy subjects. Fascicles connecting BOLD-derived clusters were analyzed within specific cortical areas: Broca s, with the pars triangularis, the pars opercularis, and the pars orbitaris; Geschwind s and Wernicke s; the premotor cortex, the dorsal supplementary motor area, the middle temporal gyrus, the dorsal prefrontal cortex and the frontopolar region. We found a functional connectome divisible into three systems-anterior, superior and inferior-around the insula, more complex than previously thought, particularly with respect to a new extended Broca s area. The extended Broca s area involves two new fascicles: the operculo-premotor fascicle comprised of well-organized U-shaped fibers that connect the pars opercularis with the premotor region; and (2) the triangulo-orbitaris system comprised of intermingled U-shaped fibers that connect the pars triangularis with the pars orbitaris. The findings enhance our understanding of language function.

We propose a novel approach to identify the foci of a neurological disorder based on anatomical and functional connectivity information. Specifically, we formulate a generative model that characterizes the network of abnormal functional connectivity emanating from the affected foci. This allows us to aggregate pairwise connectivity changes into a region-based representation of the disease. We employ the variational expectation-maximization algorithm to fit the model and subsequently identify both the afflicted regions and the differences in connectivity induced by the disorder. We demonstrate our method on a population study of schizophrenia.

The neural correlates of working memory (WM) in schizophrenia (SZ) have been extensively studied using the multisite fMRI data acquired by the Functional Biomedical Informatics Research Network (fBIRN) consortium. Although univariate and multivariate analysis methods have been variously employed to localize brain responses under differing task conditions, important hypotheses regarding the representation of mental processes in the spatio-temporal patterns of neural recruitment and the differential organization of these mental processes in patients versus controls have not been addressed in this context. This paper uses a multivariate state-space model (SSM) to analyze the differential representation and organization of mental processes of controls and patients performing the Sternberg Item Recognition Paradigm (SIRP) WM task. The SSM is able to not only predict the mental state of the subject from the data, but also yield estimates of the spatial distribution and temporal ordering of neural activity, along with estimates of the hemodynamic response. The dynamical Bayesian modeling approach used in this study was able to find significant differences between the predictability and organization of the working memory processes of SZ patients versus healthy subjects. Prediction of some stimulus types from imaging data in the SZ group was significantly lower than controls, reflecting a greater level of disorganization/heterogeneity of their mental processes. Moreover, the changes in accuracy of predicting the mental state of the subject with respect to parametric modulations, such as memory load and task duration, may have important implications on the neurocognitive models for WM processes in both SZ and healthy adults. Additionally, the SSM was used to compare the spatio-temporal patterns of mental activity across subjects, in a holistic fashion and to derive a low-dimensional representation space for the SIRP task, in which subjects were found to cluster according to their diagnosis.

This paper presents feature-based alignment (FBA), a general method for efficient and robust model-to-image alignment. Volumetric images, e.g. CT scans of the human body, are modeled probabilistically as a collage of 3D scale-invariant image features within a normalized reference space. Features are incorporated as a latent random variable and marginalized out in computing a maximum a posteriori alignment solution. The model is learned from features extracted in pre-aligned training images, then fit to features extracted from a new image to identify a globally optimal locally linear alignment solution. Novel techniques are presented for determining local feature orientation and efficiently encoding feature intensity in 3D. Experiments involving difficult magnetic resonance (MR) images of the human brain demonstrate FBA achieves alignment accuracy similar to widely-used registration methods, while requiring a fraction of the memory and computation resources and offering a more robust, globally optimal solution. Experiments on CT human body scans demonstrate FBA as an effective system for automatic human body alignment where other alignment methods break down.

This paper extends Jones’ popular electrostatic repulsion based algorithm for distribution of single-shell Q-space samples in two fundamental ways. The first alleviates the single-shell requirement enabling full Q-space sampling. Such an extension is not immediately obvious since it requires distributing samples evenly in 3 dimensions. The extension is as elegant as it is simple: Add a container volume of the desired shape having a constant charge density and a total charge equal to the negative of the sum of the moving point charges. Results for spherical and cubic charge containers are given. The second extension concerns the way distances between sample point are measured. The Q-space samples represent orientation, rather than direction and it would seem appropriate to use a metric that reflects this fact, e.g. a tensor metric. To this end we present a means to employ a generalized metric in the optimization. Minimizing the energy will result in a 3-dimensional distribution of point charges that is uniform in the terms of the specified metric. The radically different distributions generated using different metrics pinpoints a fundamental question: Is there an inherent optimal metric for Q-space sampling? Our work provides a versatile tool to explore the role of different metrics and we believe it will be an important contribution to further the continuing debate and research on the matter.

We present a novel method for inferring tissue labels in atlas-based image segmentation using Gaussian process regression. Atlas-based segmentation results in probabilistic label maps that serve as input to our method. We introduce a contour-driven prior distribution over label maps to incorporate image features of the input scan into the label inference problem. The mean function of the Gaussian process posterior distribution yields the MAP estimate of the label map and is used in the subsequent voting. We demonstrate improved segmentation accuracy when our approach is combined with two different patch-based segmentation techniques. We focus on the segmentation of parotid glands in CT scans of patients with head and neck cancer, which is important for radiation therapy planning.

Estimation of the diffusion propagator from a sparse set of diffusion MRI (dMRI) measurements is a field of active research. Sparse reconstruction methods propose to reduce scan time and are particularly suitable for scanning un-coperative patients. Recent work on reconstructing the diffusion signal from very few measurements using compressed sensing based techniques has focussed on propagator (or signal) estimation at each voxel independently. However, the goal of many neuroscience studies is to use tractography to study the pathology in white matter fiber tracts. Thus, in this work, we propose a joint framework for robust estimation of the diffusion propagator from sparse measurements while simultaneously tracing the white matter tracts. We propose to use a novel multi-tensor model of diffusion which incorporates the biexponential radial decay of the signal. Our preliminary results on in-vivo data show that the proposed method produces consistent and reliable fiber tracts from very few gradient directions while simultaneously estimating the bi-exponential decay of the diffusion propagator.

This paper investigates a diffeomorphic point-set registration based on non-stationary mixture models. The goal is to improve the non-linear registration of anatomical structures by representing each point as a general non-stationary kernel that provides information about the shape of that point. Our framework generalizes work done by others that use stationary models. We achieve this by integrating the shape at each point when calculating the point-set similarity and transforming it according to the calculated deformation. We also restrict the non-rigid transform to the space of symmetric diffeomorphisms. Our algorithm is validated in synthetic and human datasets in two different applications: fiber bundle and lung airways registration. Our results shows that non-stationary mixture models are superior to Gaussian mixture models and methods that do not take into account the shape of each point.