Abstract: The present invention discloses methods for automatically generating regions of seeding points for fiber tracking in diffusion tensor images. These methods are based on connected region grow. Seeding point selection criteria involving Fractional Anisotropy thresholding and Dominant Eigen Vector similarity are also disclosed.

Abstract: A method and system for shading large volumetric data sets using partial derivatives computed in screen-space. The method and system provide a fast and efficient shading a large datasets. Screen space derivatives are computed efficiently by evaluating neighboring pixel information together. The method can be efficiently implemented using GPUs (graphics processing units) that provides some access to information about neighboring pixels using ddx and ddy fragment processing functions.

Abstract: A system and method for segmenting an object in an image using an isoperimetric ratio in a graphics processing unit is disclosed. The object is identified by one or more selected pixels located within the object. The system includes a graphics processing unit and software loadable on the graphics processing unit. The software is operable to find weights for edges, to build a Laplacian matrix and a d vector, to eliminate the row and column corresponding to the one or more selected pixels to determine L0 and d0, to solve x0 in the equation L0x0=d0; and to threshold the potentials x at the value that selects the object having the lowest isoperimetric ratio.

Abstract: Subsets of volume data are sequentially stored for volume rendering from two dimensional textures. For example, pairs of adjacent two-dimensional images are loaded into RAM or cache. One or more strips of texture data are interpolated for polygons extending between the two-dimensional images. The strips or polygons are more orthogonal to a viewing direction than the two-dimensional images. After interpolating texture data from the two-dimensional images for a plurality of non-coplanar polygons, the texture data is rendered. The rendered information represents one portion of the three dimensional representation. Other portions are rendered by repeating the process for other pairs or subset groups of adjacent two-dimensional images. A lower cost apparatus, such as a programmed computer or a GPU with a limited amount of memory, is able to render images for three dimensional representations of very large three-dimensional arrays. The images may be rendered without copying volume data for different main axes.

Abstract: A system and method for automated and/or semi-automated analysis of video for discerning patterns of interest in video streams. In a preferred embodiment, the present invention is directed to identifying patterns of interest in indoor settings. In one aspect, the present invention deals with the change detection problem using a Markov Random Field approach where information from different sources are naturally combined with additional constraints to provide the final detection map. A slight modification is made of the regularity term within the MRF model that accounts for real-discontinuities in the observed data. The defined objective function is implemented in a multi-scale framework that decreases the computational cost and the risk of convergence to local minima. To achieve real-time performance, fast deterministic relaxation algorithms are used to perform the minimization. The crowdedness measure used is a geometric measure of occupancy that is quasi-invariant to objects translating on the platform.

Abstract: A method is provided for automatically identifying a disease outbreak indicative of a potential bio-terror attack. Patient records are mined from structured and unstructured clinical sources. The patient records are then analyzed by correlating selected patient data contained in the patient records with disease indicia for each of a plurality of diseases. A probability of a disease outbreak is estimated at least in part based on these correlations. Suspicion may also be indicated if anomalous disease clusters are found. If any of the estimated probabilities exceeds a threshold value, a disease outbreak alert is outputted. The disease indicia may be defined by disease progression models, which may be stored in a disease knowledge base.

Abstract: A method of aligning a pair of images with a first image and a second image, wherein said images comprise a plurality of intensities corresponding to a domain of points in a D-dimensional space includes identifying feature points on both images using the same criteria, computing a feature vector for each feature point, measuring a feature dissimilarity for each pair of feature vectors, wherein a first feature vector of each pair is associated with a first feature point on the first image, and a second feature vector of each pair is associated with a second feature point on the second image. A correspondence mapping for each pair of feature points is determined using the feature dissimilarity associated with each feature point pair, and an image transformation is defined to align the second image with the first image using one or more pairs of feature points that are least dissimilar.

Abstract: A method for dynamic scene modeling and change detection applicable to motion analysis utilizes optical flow for capturing and modeling the dynamics of the scene. Uncertainties in the measurements are evaluated and utilized in order to develop a robust representation of the scene in a higher dimensional space. In another embodiment, a dynamical model of the scene is developed that utilizes multiple past frames to predict the next frame. Incremental methods for updating the model are developed and, towards detection of events, a new measure is introduced that is based on a state-driven comparison between the prediction and the actual observation.

Abstract: A method and system for the classification of an object within a zone of a specified area with multiple surveillance means. The invention performs the steps of receiving a set of objects within a predefined zone area from each of at least a first and second surveillance means. Subsequently, each received set of objects is filtered to ensure that the objects in the set are comparable to the objects in the other received set. Characteristics of the received sets of objects are compared and characteristics of the objects within a received set of objects are compared to characteristics of the objects within a different set of received objects, wherein the characteristics are based upon a set of predetermined characteristics. It is determined if each object or set identified by the first surveillance means corresponds to an object or set identified by the second surveillance means.

Abstract: A method for automatic local path planning for a virtual endoscope comprises the steps of defining a sub volume around a current endoscope position in a lumen; performing a region growing inside the lumen, starting from the current endoscope position; calculating and clustering the intersection of the region with the faces of a cube circumscribing the sub volume; calculating approximated centerline paths from the current endoscope position to the center of each cluster formed in the preceding step; comparing each of the centerline paths with a current path exhibited by the endoscope; and selecting an optimal centerline path based on the comparison.

Abstract: An apparatus according to an embodiment of the present invention is provided for sensing acoustic signals. The apparatus includes a housing having an apertured posterior member, a sensing unit extending through the apertured posterior member for interfacing by contact with a patient, a cursor control for positioning a cursor on a display, the display located anteriorially on the housing, a button for fixing a position of the cursor in the display, indicating the position on the housing with respect to the patient, and a circuit for causing the cursor to move about the display in response to the cursor control, and for transmitting the acoustic signal.

Abstract: A client-server system and method that enables efficient, low bit rate transmission of image data over a network from an image server (e.g., active cameras) to a client for, e.g., distributed surveillance. A detection and tracking module detects a human presence within an observation area and provides 2-dimensional face coordinates and its estimated scale to a video transmission module. The captured video is then efficiently encoded in log-polar coordinates using an adaptive log-polar mapping, with a foveation point centered on the face. A fovea region, which covers the target object (face), is uniformly sampled and transmitted at full resolution to the client. The periphery region (background) is sampled according to a log-polar grid.

Abstract: The present invention provides techniques for combining various types of diagnostic images. In one embodiment of the present invention, a composite image is obtained by fusing MPR and MIP images. The MPR is obtained from a CT scan, and the MIP is obtained from a PET scan. The resulting superimposed images help a physician to see the diagnostic information in context. In other embodiments of the present invention, techniques for an MIP-MIP overlay, VR-MIP overlay, and VR-MPR overlay are provided.

Abstract: A system for generating a diagnosis of the probable cause of detected faults. Preferably, a user interface is provided and a Bayesien network is employed wherein probabilities are generated automatically and a manual process is used to construct a probability table. The system also allows multiple hypotheses and/or diagnoses to be presented simultaneously to the operator.

Abstract: A three-dimensional (3D) imaging system and method include a processor for generating a volume-rendered 3D image on a display using a plurality of voxels from a 3D image dataset; and a view selector which responds to user inputs for determining a first set of voxels corresponding to a boundary in the volume-rendered 3D image using a predetermined boundary-specifying criteria applied to a function of the opacity of the plurality of voxels. The processor responds to the first set and to a user function selection by performing the selected function to modify the volume-rendered 3D image relative to the detected boundaries. The selected function is a display function, a measurement function, or a segmentation function. The view selector determines the boundary from voxels having opacities greater than a predetermined threshold, or determines the boundary from voxels having a gradient in opacities greater than a predetermined threshold.

Abstract: Improved image reconstruction methods for cone beam ROI (region of interest) imaging of long objects with an area detector using a two or more circular scans and a connecting line scan, wherein essentially the entire detector area is utilized for acquiring cone beam data for all source positions.

Abstract: A method for automatic CR image composition includes the following steps for each image in a CR image pair. A vertical gradient image is projected onto a y-axis to obtain a y-axis projection. Edge positions (candidates) are identified from white band edge positions based on an absolute maximum value of the y-axis projection. An intensity change constancy identifies candidates having orientation angles less than a threshold angle with respect to a horizontal. An intensity change value verifies intensity value differences on two candidate sides. A cross-correlation score is computed for the candidates having an error value below a threshold for the intensity change constancy and the intensity change value, by comparing a consistency of the candidates against image data of the CR image pair. A final overlap is identified for the CR image pair, based upon the cross-correlation score.

Abstract: A method for matching images includes the step of providing a template image and an input image. A Laplacian-of-Gaussian filtered log (LOG-log) image function is computed with respect to the template image and the input image to obtain a Laplacian-of-Gaussian filtered template image and a Laplacian-of-Gaussian filtered input image, respectively. An energy function is minimized to determine estimated geometric transformation parameters and estimated photometric parameters for the input image with respect to the template image. The energy function is formed by weighting non-linear least squared differences of data constraints corresponding to locations of both the Laplacian-of-Gaussian filtered template image and the Laplacian-of-Gaussian filtered input image. The estimated geometric transformation parameters and the estimated photometric parameters are output for further processing. The method allows for image matching under non-uniform illumination variations.

Abstract: A system and method for detecting pedicle positions in an image, in accordance with the present invention, includes providing a set of feature prototypes for a plurality of pedicle positions and orientations, providing an input image to be analyzed for pedicle positions and orientations, and determining intensity curvatures for a pedicle in the input image. The intensity curvatures are transformed to determine a feature vector for the pedicle in the input image. The feature vector is correlated to the feature prototypes to determine most likely positions and orientations of the pedicle.

Abstract: A method for tracking a video object in a time-ordered sequence of image frames, comprises the steps of (a) identifying the object to be tracked in a first frame and determining its shape in the first frame; (b) selecting another frame, referred to as the last frame and identifying the shape of the object in the last frame; (c) selecting a number of intermediate frames between the first and last frames; (d) automatically identifying the shape of the object in the intermediate frames; and (e) interpolating between all of the frames in the sequence so as to determine the trajectory of the object in all of the frames.