X-ray or gamma ray systems or devices – Specific application – Computerized tomography
Reexamination Certificate
2000-06-08
2001-10-02
Bruce, David V. (Department: 2882)
X-ray or gamma ray systems or devices
Specific application
Computerized tomography
C378S017000, C378S901000
Reexamination Certificate
active
06298110
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention is directed to a method of and system for computed tomography (CT) density image reconstruction. More particularly, the present invention is directed to the three-dimensional reconstruction from two-dimensional projections acquired with x-ray cone-beam CT and single photon emission computed tomography (SPECT) scanners. Even more particularly, the present invention is directed to a method of and system for intravenous volume tomographic digital angiography imaging.
For about the past twenty years, computerized tomography has revolutionized diagnostic imaging systems as well as non-destructive test imaging techniques. Conventional CT scanners use a fan-shaped x-ray beam and one-dimensional detector in order to reconstruct a single slice with a single scan of an object However, current CT technology is limited by a trade-off between high longitudinal resolution and fast volume scanning. One method which has been utilized to address the shortcomings of CT scanner technology is the use of cone-beam tomography. A cone-beam volume CT scanner uses a cone-beam x-ray source and a two-dimensional detector to reconstruct the whole volume of an object with a single scan of that object The data obtained from the scan of the object is processed in order to construct an image that presents a two-dimensional slice taken through the object. The current technique for reconstructing an image from 2-D is referred to in the art as the filtered back projection technique. That process converts the attenuation measurements from a scan into integers called “CT numbers” or “Hounsfield units” which are then used to control the brightness of a corresponding pixel on a cathode ray display.
In a 3-D scan technique, a cone-shaped x-ray beam is used which diverges to form a cone-beam that passes through the object and impinges on a two-dimensional array of detector elements. In that manner, the volume scanning time of a 3-D object can be at least 10 times shorter than a standard CT on a spiral CT. In contrast to existing CT with an intraslice plane resolution of 1.0 lp/mm, the reconstructions of cone beam CT will have isotropic spatial resolution along all three axes (0.5-2.0 lp/mm). Each view is thus a 2-D array of x-ray attenuation measurements and the complete scan produces a 3-D array of attenuation measurements.
At present, either of two methods are commonly used to reconstruct a set of images from the acquired 2-D attenuation measurements. The first technique is that developed by Feldkamp, Davis & Kress, which is described in “Practical Cone-Beam Algorinthm”,
J. Opt. Soc. Am
., Vol. I, pp. 612-619 (1984). The Feldkamp, et al. technique, which uses an algorithm which was derived using approximations of a tilted fan beam formula, is a convolution-back projection method which operates directly on the line integrals of the actual attenuation measurements. That method can be easily implemented with currently available hardware and is a good reconstruction for images at the center or “mid-plane” of the cone-beam. While the algorithm of Feldkamp, et al. provides excellent computational efficiency and minimal mechanical complexity in data acquisition, its major shortcoming is that it is based on single circle cone-beam geometry. Single circle cone-beam geometry, in which the source always lies on a circle, cannot provide a complete set of data to exactly reconstruct the object. For that reason, Feldkamp, et al.'s algorithm causes some unavoidable distortion in the non-central transverse planes, as well as resolution degradation in the longitudinal direction.
In order to address the problems of Feldkamp's algorithm, several other approaches have been proposed using different cone-beam geometries including dual orthogonal circles, helical orbit, orthogonal circle-and-line, and Smith's curve. Such geometries can achieve exact reconstructions when using the approach of Tuy, Smith, or Gangreat.
In addition to the Feldkamp, et al. approach for analytic cone-beam reconstruction, a second commonly used method is that disclosed by Pierre Grangeat in, “Mathematical Framework of Cone-Beam 3-D Reconstruction Via the First Derivative of the Radon Transform”,
Mathematical Methods in Tomography
, Herman, Lewis, Natterer (eds.) Lecture Notes in Mathematics, No. 1497, pp. 66-97, Spring Verlag (1991). That algorithun provides an accurate solution to the image reconstruction task based on a fundamental relationship between the derivative of the cone-beam plane integral through the derivative of the parallel beam plane integral. While the Grangeat method is theoretically accurate, it requires mathematical operations that can be solved only using finite numerical calculations that are approximations. Thus, errors can be introduced by the implementation of the Gangreat method that can be greater than those produced using the Feldkamp, et al. method and such errors are not correlated with the cone-beam angle. A third method has been disclosed by H. K. Tuy in “An Inversion Formula for a Cone-Beam Reconstruction”, SAIM J. Appl. Math. 43, pp. 546-552 (1983). Using Tuy's approach, in order to generate a complete or sufficient set of data, every plane which passes through the imaging field of view must also cut through the orbit of the focal point at least once. The single plane or orbit of Feldkamp, et al. does not satisfy this condition.
Still yet another approach that has been proposed is the inversion of the cone-beam data sets if the assumption is made that for any line that contains a vertex point and a reconstruction point, there is an integer M which remains constant for the line such that almost every plane that contains this line intersects the geometry exactly M times. Mathematical improvement to the reconstruction algorithms was described in an article by B. D. Smith entitled “Cone-Beam Tomography: Recent Advances and a Tutorial Review,”
Opt. Eng
., Vol. 29 (5) pp. 524-534 (1990). However, such an integer requirement condition is too restrictive for practical application since the only known source point geometry which meets that condition is a straight line.
Two somewhat recent patents were issued in the United States directed to the cone-beam reconstruction problem. The first, U.S. Pat. No. 5,170,439 to Zeng, et al., was issued on Dec. 8, 1992 and utilizes the above-described cone-beam reconstruction method using combined circle and line orbits. However, that technique requires the removal of redundant and unnecessary which necessarily requires more computing time and complexity than the method and system of the present invention.
Another approach to the reconstruction of images from cone-beam data is disclosed in U.S. Pat. No. 5,400,255, which issued to Hu on Mar. 21, 1995. The methodology disclosed in the Hu patent represents a minimal improvement from Feldkamp's algorithm and it is still an approximate method based on a single circle cone beam geometry. It cannot result in exact reconstruction and it is not acceptable in many clinical applications when the cone angle is large.
In contrast to the prior art approaches, the present invention discloses an exact cone-beam reconstruction system and method using a circle-plus-arc data acquisition geometry in which the locus of a source and a detector is a circle plus an orthogonal arc. In that manner, the best image quality of a cone-beam volume CT is achieved without introducing any additional mechanical complexity compared to a regular CT gantry. If the locus of an x-ray source and a detector is a single circle during cone-beam scanning (single circle cone-beam geometry), an incomplete set of projection data will be acquired. The incompleteness of the projection data results in some unavoidable blurring in the planes away from the central z plane and a resolution loss in the z direction (i.e., Feldkamp, et al.'s algorithm). The reconstruction error due to the incompleteness of the projection data could be up to 50% of the signal when using Feldkamp, et al.'s algorithm with a 22° cone angle
Blank Rome Comisky & McCauley LLP
Bruce David V.
University of Rochester
LandOfFree
Cone beam volume CT angiography imaging system and method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Cone beam volume CT angiography imaging system and method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Cone beam volume CT angiography imaging system and method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2580490