Abstract: A method and apparatus for three dimensional (3D) computerized tomographic (CT) imaging of a region-of-interest (ROI) in an object, wherein image reconstruction processing is applied to a plurality of sets of 2D cone beam projection data, each set being acquired on a 2D detector at a corresponding plurality of scan path source positions. A first image reconstruction processing step comprises applying a mask to each set of the projection data so that data inside the boundaries of each mask form a corresponding plurality of masked 2D data sets. Next, the data inside each masked 2D data set is filtered along a plurality of parallel lines formed therein, to generate a corresponding plurality of filtered 2D data sets. Each filtered 2D data set corresponds to a calculation of a first estimate of Radon derivative data determined from a given set of the 2D cone beam projection data. The next step comprises adaptively developing 2D correction data for each of the first estimates of Radon derivative data.

Abstract: A scanning and data acquisition method and apparatus for three-dimensional computerized tomographic imaging of a region of interest (ROI) of an object which is smaller than the object itself and having upper and lower boundaries which are completely within a field of view of an imaging system comprises merely a continuation of the scan trajectory used for scanning the main portion of the ROI so as to extend past its upper and lower boundaries. In a preferred embodiment, the scan path of the present invention consists of a main spiral scan path comprising a plurality of spiral turns, or stages, for scanning between upper and lower boundaries of the ROI, and at least a portion of an extra single spiral turn of the scan path at each end thereof.

Abstract: In accordance with the principles of the present invention, image reconstruction errors which occur when using an area detector having a plurality of rows of detector elements defining a height for the detector that is smaller than the cone beam image, can be avoided by correlating an amount of orthogonal translation applied to line segments L extending across the detector towards top and bottom edges thereof that are used for calculating Radon derivative data, with the spacing between adjacent rows of the detector elements that are at the top and bottom edges of the detector. In a preferred embodiment the detector comprises a plurality of M rows of detector elements centered between the top and bottom edges of the detector, and an additional N rows of detector elements adjacent the M rows, at both of the top and bottom edges of the detector (where N may equal 1).

Abstract: In an arrangement for computed tomography imaging, line integral data is converted into a set of Radon data, for use in constructing an image of an object, by computing the derivatives for respective data points in the Radon data set. To improve efficiency in derivative computation, the Radon space is partitioned by a set of concentric spherical shells, so that each of the Radon data points lies on one of the shells. An equal number of spherical shells corresponds to each processor in an array of processors, and each processor is operated to compute respective Radon derivatives only for Radon data points lying on its corresponding shells.

Abstract: An arrangement for CT imaging of an object is provided, wherein a cone-beam X-ray source traverses a scan path to acquire line integral data at a plurality of view positions, a set of Radon data points in Radon space being associated with the arrangement, and it being necessary to compute the Radon derivative for each Radon data point in order to construct an image of the object. The arrangement includes a specified number of processors greater than one. The Radon space is partitioned by a number of coaxial planes separated from one another by an angle of .pi. divided by the specified number. A set of view positions likewise separated from one another by an angle equal to .pi. divided by the specified number are selected from the plurality of view positions, and each view position in the view position set is projected into a different one of the coaxial planes to define a subset of Radon data points corresponding to one of the planes, and also to one of the processors.

Abstract: A system is provided for improving the operation of a computed tomography system in forming an image of an object, wherein a cone-beam source is employed to project an image of the object onto a detector plane to provide a set of cone-beam projection data, and such data is to be converted into a set of Radon data for use in constructing the desired image. In the invention, a line L' is rotated in the detector plane from a line L about a center of rotation which lies on the line L, so that L' lies at a small angle with respect to line L. Data from the cone-beam projection data set is respectively integrated along the first and second lines to generate first and second weighted line integrals respectively corresponding to the lines L and L' The two weighted line integrals are then employed to find the derivative of one of the data points in the Radon data set.

Abstract: The present invention discloses a method and system for performing three-dimensional computerized tomography imaging of a region of interest of an object. In the present invention, a scanning trajectory is provided about the object. The scanning trajectory includes a first scanning circle, a second scanning circle, and a helical scanning path connecting the first scanning circle and the second scanning circle. The scanning trajectory is then sampled with a plurality of cone beam source positions. Cone beam energy is emitted along the scanning trajectory from each of the plurality of cone beam source positions towards a portion of the object. Cone beam energy passing through the object is acquired on a detector as cone beam projection data. The cone beam projection data is then masked with a plurality of masks. Each of the masks remove cone beam projection data that is outside the region of interest of the object and retain cone beam projection data that is within the region of interest of the object.

Abstract: The present invention discloses a method and system for generating Radon derivative data in a cone beam computerized tomography implementation. Radon derivative data is computed by using a plurality of processors that partition the Radon space among the processors in a manner wherein the totality of the support of the Radon space handled by each processor is approximately equal. More specifically, each processor handles a number of vertical planes in the Radon space which are equally spaced within 180.degree. in the .phi..sub.k orientation. This procedure enables the plurality of processor to equalize the workload in computing the Radon derivative data and to perform the computations in a timely and efficient manner.

Abstract: The present invention discloses a method and apparatus for generating Radon derivative data on locations proximate to a near uniformly spaced polar grid in a digitized Radon space for a cone beam computerized tomography (CT) implementation. Radon derivative data is determined for uniformly spaced coordinates in r, .theta. and .phi. directions.

Abstract: Imaging of a region of interest within a larger object is accomplished without the need for determining Radon derivatives of portions of the object outside a field of view which generally corresponds to the region of interest. The field of view and region of interest may be relatively large compared to a relatively small area detector used for the imaging. In order to provide a complete data set satisfying Radon completeness requirements with little or no collection of data from outside the region of interest, a source scanning trajectory uses a first circle, a second circle, and a helical portion connecting the first and second circles. The first and second circle and helical portion define a cylinder which is outside and surrounding the field of view, which is likewise a cylinder.

Abstract: Construction and assessment methods provide a three dimensional scanning trajectory to ensure the acquisition of a complete set of Radon data for exact image reconstruction of an object irradiated by a cone beam source is described. The three dimensional scanning trajectory must be nowhere disconnected permitting the source to traverse the scanning trajectory in a continuous manner from start point to end point. The trajectory defines edges wherein a convex surface is formed by connecting these edges, this surface is known as a convex hull. If the convex hull defined by a trajectory encloses the object to be scanned, a complete set of Radon data for exact image reconstruction of the object can be acquired. This provision amounts to satisfying the three dimensional completeness criterion. Furthermore, if any planar projection of the trajectory encloses the corresponding planar projection of the object, the three dimensional completeness criterion is satisfied by using projected planar visualization aids.

Abstract: Techniques and system for pre-processing cone beam projection data for reconstructing substantially free of interpolation-induced artifacts a three-dimensional computerized tomography (CT) image of a portion of an object are provided. Such techniques include identifying first, second and third regions on a surface array detector such that only cone beam projection data acquired in the identified regions is retained for subsequent processing. The identified regions cooperate to eliminate interpolation effects upon lines of integration situated relatively close to boundaries on the array detector and thus allow for reconstructing the substantially free of interpolation-induced artifacts CT image.

Abstract: Computerized tomography (CT) reconstruction of images is accomplished for asymmetric grids. After initial data acquisition and normalization, the asymmetric projection data is distorted by a compression process which effectively converts the data into symmetric data. The data may then be processed using Radon reconstruction procedures which convert projection data into object density functions. The object density functions or data which are provided are then subject to a stretching or extension procedure which compensates or reverses the initial distortion or compression. Cone beam imaging with asymmetric grids may be two-dimensional or three dimensional when utilizing the technique.

Abstract: Computerized tomography provides three-dimensional imaging by applying cone beam energy to an object to be imaged. The energy passes through the object and is detected. The resulting cone beam data is efficiently processed to provide Radon data on polar grid points on a series of vertical planes. The Radon data is calculated by initially defining a Radon circle on each of the coaxial vertical planes. Radon derivative data is calculated at intersection points between the Radon circle and the radial lines. The values of Radon derivative data are binned according to the nearest grid point or nearest two grid points. Values are stored in the bins for each point in a particular plane until values for all of the planes have been supplied. The values for all the points are recalculated with each of a series of source positions, corresponding to the relative positioning of the source of cone beam energy and the object being imaged.

Abstract: Techniques and system for processing cone beam projection data for reconstructing substantially free of boundary-induced artifacts a three-dimensional computerized tomography (CT) image of a portion of an object are provided. Such techniques include suitably identifying a rotation center shared by a line of integration pair and wherein the rotation center is selected for mapping, within a cone beam masked region identified on a surface array detector, predetermined points situated along the line of integration pair. The suitably identified rotation center allows to acquire cone beam projection data within the masked region which is free of boundary effects. The acquired data is retained for subsequent processing and thus allows for exactly reconstructing the substantially free of boundary-induced artifacts CT image.

Abstract: Complete cone beam scanning and data acquisition techniques are provided for three-dimensional computerized tomography imaging of an object. Such techniques include defining a source scanning trajectory such as a helical path located on a cylindrical surface which surrounds a field of view that contains the object. The foregoing scanning trajectory allows using area detectors having a predetermined dimension (such as height or width) substantially smaller than the corresponding dimension of the object to be imaged.

Abstract: An improved method and apparatus for preprocessing cone beam attenuation data to reconstruct a three dimensional image of a particular region of interest of an object by a process of inverse Radon transformation is described. The number of required operations is substantially reduced by selectively retaining for further processing only that cone beam attenuation data acquired within a select closed region of the surface of the array detector wherein this region provides data corresponding to beams actually attenuated in passing through the region of interest of the object. In this manner, unnecessary beam attenuation data is discarded at the earliest possible opportunity to expedite image processing.

Abstract: A method and apparatus for exact three dimensional reconstruction of the image of a portion of interest of an object in the field of view of a cone beam source is disclosed, whereby unwanted Radon data are selectively disgarded avoiding undue corruption of the imaging process and missing Radon data that would otherwise be lost are selectively recovered. This is accomplished by ensuring the acquisition of a complete set of Radon data through proper choice of scanning configuration, then selectively partitioning and manipulating the acquired data to image only the select portion of interest of the object.

Abstract: A method and apparatus for acquiring scanning cone beam projection data along a three dimensional scanning trajectory is disclosed. The technique ensures acquisition of a substantially uniform distribution of Radon data that is sufficiently complete for exact image reconstruction of an object irradiated by a scanning cone beam source. By choosing a scanning trajectory whose planar projection onto a plurality of coaxial planes, when taken together, forms a continuous, convex, closed curve about a corresponding projection of the object being scanned; the Radon information obtained therefrom is complete and information coverage is uniform over the region of support of the object. Accordingly, this condition provides a practical, easy to implement, visual method for evaluating candidate three dimensional scanning trajectories.

Abstract: A multi-resolution array detector includes a fine resolution zone having a higher density of detector elements than in a coarse resolution zone having a relatively lower density of detector elements. The array detector, usually an area detector, is used in computerized tomography (CT) imaging to provide a finer resolution image in one zone than in another zone. A method of processing data from the multi-resolution detector uses a non-uniform distribution of data in Radon space.