Abstract: A virtual internal cavity inspection system non-invasively provides images of cavities of a subject from a viewpoint within the cavity. An acquisition unit acquires imaging information about internal structures of a subject. This imaging information is segmented into separate structures, and a 3D surface model is constructed. An operator views an image of the 3D model to select a goal viewpoint and a start viewpoint. A viewpoint path is created linking the start viewpoint and the goal viewpoint. Images are created with a viewpoints along the viewpoint path. A variety of additional visualization techniques aid the viewer's localization of the current image viewpoint, and its relation to the subject.

Abstract: A method of determining surfaces of swept volumes, which defines a region reserved for the removal of an object, or the motion of an object, is determined employing implicit modeling. A definition of an object and the trajectory in which it is to be moved are provided to the swept surface display device. An implicit model is created by determining shortest distance from each voxel of a object volume to a surface point of the object. A workspace volume has voxels each initialized with distances which are much larger than any distance envisioned. The implicit model space voxels are transformed relative to the workspace voxels according to the trajectory at a time t. Workspace voxels are updated with corresponding implicit model space voxels when the value of the implicit model space voxel is lower than the workspace voxel value. The implicit model space voxels are transformed relative to the workspace voxels for another time t and the number of workspace voxels are again updated.

Abstract: Three-dimensional (3D) image data is acquired from a subject with a medical imaging device and stored. The stored 3D image data is processed by a model workstation to segment the model into discrete structures and produce a segmented computer graphic model. An operator interacts with the model workstation to cause it to display desired structures of the the segmented model in a desired view and orientation. The operator also selects a position and orientation of a cutting plane passing through the segmented model. Once selected, the position and orientation information is provided to a pulse sequencer of a magnetic resonance (MR) imaging system. The pulse sequencer controls an RF transmitter and gradient amplifiers to cause an MR image of the subject at an imaging plane, corresponding to the cutting plane of the model workstation, to be acquired.

Abstract: A non-invasive imaging device obtains a four dimensional (4D) image data set of a living subject representing three dimensions and time in relation to the subject's cardiac cycle. In order to determine various vascular parameters, it is useful to segment the image data set into internal structures defined as having the same tissue types contiguous locations. To accomplish this, a gradient calculation unit constructs a gradient data set from the image data set indicating the magnitude of spatial changes in the image data set. A plurality of locations are selected by an operator with a pointing device during `training` along with corresponding data values in the image and gradient data sets. These data values are plotted by a scatter generator against each other to construct a scatter plot then processed to determine a bivariate statistical probability distribution. The remaining data values are then assigned a tissue type based upon their plot on the bivariate statistical probability distribution.

Abstract: A 4D image data set of a living subject is obtained by non-invasive imaging means representing three dimensions and time in relation to the subject's cardiac cycle. In order to determine various vascular parameters, it is useful to segment the image data set into internal structures defined as having the same tissue types contiguous locations. To accomplish this, a gradient data set is constructed from the image data set indicating the magnitude of spatial changes in the image data set. A plurality of locations are selected in `training` along with corresponding data values in the image and gradient data sets. These data values are plotted against each other to construct a scatter plot then processed to determine a bivariate statistical probability distribution. The remaining data values are then assigned a tissue type based upon their plot on the bivariate statistical probability distribution. Contiguous locations having the same tissue type assignment are identified as a solid structure.

Abstract: In order to display the surfaces of internal structures within a solid body from non-intrusively acquired data sets, it is useful to segment the data sets into the internal structures of interest before searching for the surfaces of such structures. To accomplish this, a data segmentation system uses a plurality of sample data points to construct a statistical probability distribution for a plurality of internal structures. Using these probability distributions, each data point is labeled with the most likely structure identification. Searching the thus-segmented data points for surfaces is considerably faster than is possible with the entire data set and produces surface renditions with fewer anomalies and errors. A non-intrusive imaging means is used to obtain a 3D data set. The probability distribution is bivariate and the two data sets are plotted against each other to assist in identifying tissue types.

Abstract: A magnetic resonance (MR) imaging system for use in a medical procedure employs an open main magnet allowing access to a portion of a patient within an imaging volume, for producing a main magnetic field over the imaging volume; a set of open gradient coils which provide magnetic fields gradients over the imaging volume without restricting access to the imaging volume; a radiofrequency coil set for transmitting RF energy into the imaging volume to nutate nuclear spins within the imaging volume and receive an MR response signal from the nuclear spins; and a pointing device for indicating the position and orientation of a plane in which an image is to be acquired; an image control means for operating power supplies for the gradient coils and the RF coils to acquire an MR signal from the desired imaging plane; and a computation unit for constructing an image of the desired imaging plane.

Abstract: A method and apparatus for converting computed tomography (CT) data into finite element models. The system generates both 2-D and 3-D models using the automatic mesh generators, QUADTREE and OCTREE, which are founded on recursive spatial decomposition. Multiple slices of CT data are obtained by scanning the object to be modeled. The slices are stacked and processed to form a discrete solid model. The discrete solid model is an alternate geometry defined discretely rather than with continuous analytic curves and surfaces but still provides a foundation for automatic mesh generation. Since the QUADTREE and OCTREE automatic mesh generators map naturally to the discrete solid model, the integration of CT technology and automatic mesh generation can be achieved.

Abstract: A three dimensional image of a human brain or other body structure is constructed using a single flow sensitive data array and a flow insensitive data array to generate the contrasts necessary to differentiate among stationary tissues and also between stationary tissues and flowing blood. A plurality of data points from this combined image data are identified to tissue types and used to segregate the remaining data by using a nearest neighbor process in which each data value takes the tissue type of its nearest neighbor data point.

Abstract: In order to display the surfaces of internal structures within a solid body from non-intrusively acquired data sets, it is useful to segment the data sets into the internal structures of interest before searching for the surfaces of such structures. To accomplish this, a data segmentation system uses a plurality of sample data points to construct a statistical probability distribution for a plurality of internal structures. Using these probability distributions, each data point is labeled with the most likely structure identification. Searching the thus-segmented data points for surfaces is considerably faster than is possible with the entire data set and produces surface renditions with fewer anomalies and errors. If the solid body is a human head, and nuclear magnetic resonance (NMR) is used to obtain two data sets corresponding to the two NMR echoes, then the probability distribution is bivariate and the two echoes can be plotted against each other to assist in identifying tissue clusters.

Abstract: In a method and apparatus for detecting and displaying arbitrary interior surfaces of a three-dimensional body from a regular array of values of at least one physical property in the interior of the body, the physical property measurements are made with systems such as computerized tomographic x-ray imaging, or magnetic resonance imaging. A recursive algorithm starts with a seed voxel in the surface of interest and continues to all adjacent voxels having a common face which is penetrated by the surface of interest. The common surface penetrations are derived from a look-up table which lists the adjacent voxels having a common penetrated face for each voxel index. The voxel index, in turn, is the string of binary digits representing the voxel vertices whose vertex values exceed the surface value of the surface of interest.

Abstract: The present user interface for a golf green and golf putt modelling system includes actuators which allow a user to select camera positions and illumination of a golf green. The user also may select, using the interface, a location of a golf ball and a golf cup on the green. The interface also displays the path trajectories of the golf ball on the green.

Abstract: A method and apparatus for displaying arbitrary cross-sectional views of a three-dimensional body from a regular array of values of at least one physical property in the interior of the body includes making physical property measurements with such systems as computerized tomographic x-ray systems, or magnetic resonance imaging systems. Cut planes are defined by the user as displacement from, and rotations from, an arbitrary coordinate origin in the data space. An initial plane of pixel positions of arbitrary density is displaced and rotated to correspond to the cut plane. The values for the physical property at the pixel positions are interpolated from the surrounding measurements of actual values. Cross-sectional images are thereby supplied interactively in real time to support, for example, ongoing surgical procedures.

Abstract: A count is kept of the number of surfaces traversed, in a variety of three-dimensional imaging systems which are particularly advantageous for use in medical applications based upon data generated by magnetic resonance imaging systems and computerized axial tomographic imaging systems. The system is particularly advantageous in a surgical workstation for planning a wide variety of operative procedures. Surfaces may be selected and displayed simultaneously so as to provide the viewer with a significantly improved knowledge of the relationship of various internal body structures.

Abstract: Data in tomographic images is convolved with a low-pass filter to reduce the maximum spatial frequency permitting unique mapping of surface images to a rasterized display. The filter may use linear or other weighting. Re-sampling after the filter is optional. The filtered tomographic data is employed to determine the locations and normal vectors of the surface. The locations and normal vectors employed by a computer graphics processor for applying shading in relation to the angle between the normal vector to a surface element and an operator- defined line of sight to the surface.

Abstract: A dual pipeline architecture for a system for preprocessing image data gathered from such systems as magnetic resonance imaging and computer axial tomography scanners. In one simultaneously operable dual pipeline, normal vectors are computed for each voxel element in accordance with the dividing cubes vector generation system. In another pipeline, circuitry is provided for generating surface identification indicia which are associated with unit normal vectors produced by the former pipeline. Data are supplied to a graphics processor for the production of shaded two-dimensional images representative of three-dimensional surfaces from various view angles. The system is particularly useful in medical diagnostic applications and is particularly useful in planning surgical procedures.

Abstract: An apparatus and method for determining connected substructures within a body is disclosed. The system and method are particularly advantageous for use in medical diagnostic imaging applications. In particular, three dimensional regions exhibiting the same tissue type are similarly labeled. The system and method of the present invention is operable in any situation in which three dimensional signal patterns representing the value of one or more physical properties associated with the substructures within a body at regularly spaced grid locations within the body are presented. A subset of the original data may then be provided to a display processor, particularly one employing gradient normal shading for display of three dimensional images.

Abstract: An apparatus and method for determining connected substructures within a body is disclosed. The system and method are particularly advantageous for use in medical diagnostic imaging applications. In particular, three dimensional regions exhibiting the same tissue type are similarly labeled. Using the label information together with a seed location located in the substructure of interest, all similarly labeled connected data points are determined. The system and method of the present invention is operable in any situation in which three dimensional signal patterns representing the value of one or more physical properties associated with the substructures within a body at regularly grid locations within the body are presented. A subset of the original data is then provided to a display processor particularly one employing gradient normal shading for display of three dimensional images.

Abstract: A method and apparatus for displaying three dimensional surface images includes the utilization of a case table for rapid retrieval of surface approximation information. Eight cubically adjacent data points associated with a given voxel element are compared with a predetermined threshold value or range to generate an eight bit vector. This eight bit vector is employed to rapidly produce vector lists of approximating surfaces. A non-linear interpolation operation is performed so as to more closely approximate the desired surface and to provide more accurate representations of vectors normal to the desired surface. The accurate representation of these normal directions provides means for accurately representing shading information on a display screen. The method and apparatus of the present invention are particularly useful for the display of medical images both from X-ray generated data and from data generated from various other sources including magnetic resonance imaging and positron emission tomography.

Abstract: A system for displaying three dimensional surface structures employs the analysis of voxel elements defined by eight cubically adjacent grid locations which are associated with at least one physical property defined throughout a three dimensional body. The physical property measurements may be made with such systems as computerized tomographic x-ray systems, or magnetic resonance imaging devices. Surface structures are defined by a user selected threshhold value such as distinct values for skin and bone in medical diagnostic applications. Selected voxels are subdivided and means are provided for generating signal values associated with intermediate additional grid locations. Means are also provided for generating surface normal vectors associated with these grid locations.