Image analysis – Applications – 3-d or stereo imaging analysis
Patent
1997-08-21
1999-07-27
Chang, Jon
Image analysis
Applications
3-d or stereo imaging analysis
382131, 345419, 25036304, 378 4, 600425, G06K 900, A61B 600
Patent
active
059303840
DESCRIPTION:
BRIEF SUMMARY
DESCRIPTION
1. Field of the Invention
The invention relates to a process for the reconstruction of a good resolution, contrasted, three-dimensional image of an object defined by values assumed by a property in an array of elementary volumes of the 3D image of the object.
The invention also relates to an application of this process to the production of an attenuation cartography or mapping of the object, the property of the object then being the attenuation of radiation by said object.
The invention has numerous applications in the field of imaging and in particular medical imaging, especially for the study of the vascular tree in X imaging on the basis of an attenuation cartography of said vascular tree.
2. Prior Art
In the field of three-dimensional (3D) reconstruction of images, the expert frequently uses algebraic or analytical resolution methods.
Among the analytical methods, the most frequently used algorithm is the filtering/reprojection algorithm described by NATERRER in "The mathematics of computerized tomography", J. WILEY and Sons, 1986. This algorithm is also considered to be the reference algorithm in this field.
In an application to attenuation cartography to be described in greater detail hereinafter, said algorithm is essentially used when the chosen acquisition geometry is a parallel or fan-type geometry.
However, for a cone-beam acquisition geometry, it is e.g. possible to use: cone-beam geometry described by FELKAMP C. A. et al in "Practical cone-beam algorithm", 1984, J. Opt. Soc. Am.; the HILBERT transform of the primary derivative of the Radon transform, described by SMITH B. D. in "Image reconstruction from cone-beam projections : necessary and sufficient conditions and reconstruction methods", 1985, IEEE Trans. on Med. Imag. and by KUDO H., SAITO T. in "3D tomographic image reconstruction from incomplete cone-beam projections", Proc. Topical Meeting OSA, Signal Recovery and Synthesis, Cape Code (USA), 1989; the primary derivative of the Radon transform, described by GRANGEAT in "Analysis of a 3D imaging system by reconstruction on the basis of X-radiography in cone-beam geometry", 1987, PhD thesis submitted to the Ecole Nationale Superieure des Telecommunications (Paris) and also described in European patent application EP-A-29240.
Although these algorithms permit a rapid processing, they suffer from the major disadvantage of being unusable when the acquisition conditions are difficult, i.e. for example in cases where the angular discretization of the acquisition is not regular or when the angular sector or number of projections is very small or in the case where the acquisition geometry is not stable, which creates fuzzy measurements.
Algebraic methods generally use measured projections of the property of the object and then, on the basis of these measured projections, reconstruct a first 3D image of the object. It is then possible to deduce said first 3D image from the projections (called calculated projections of the object), which are compared with real projections of the object. The result of this comparison between the real projections and the calculated projections makes it possible to update the reconstructed 3D image. On the basis of this new 3D image, new calculated projections are deduced and are again compared with the real projections. The process continues in iterative manner until a satisfactory 3D image is obtained.
For a better general understanding of this known algebraic reconstruction process, reference can be made to FIG. 1 showing a general diagram for the reconstruction of a 3D image by a conventional algebraic method.
In FIG. 1, PM refers to the real projections determined for the studied property and PC designates the projections calculated on the basis of the reconstructed 3D image. The reconstructed 3D image is called I3D. Moreover, E1 is the stage of comparing the real projections with the calculated projections, E2 the stage of updating the 3D image and E3 the reprojection stage for deducing new calculated projections.
Thus, such an algebraic method is i
REFERENCES:
patent: 4095107 (1978-06-01), Genna et al.
patent: 4769756 (1988-09-01), Webber et al.
patent: 5321612 (1994-06-01), Stewart
patent: 5390291 (1995-02-01), Ohashi
patent: 5400255 (1995-03-01), Hu
patent: 5408511 (1995-04-01), Grangeat et al.
patent: 5430783 (1995-07-01), Hu et al.
patent: 5475726 (1995-12-01), Azevedo et al.
patent: 5559335 (1996-09-01), Zeng et al.
patent: 5560360 (1996-10-01), Filler et al.
patent: 5661773 (1997-08-01), Swerdloff et al.
R.J. Michellena, "Singular-Value Decomposition for Cross-Well Tomography," Geophysics --vol. 58(11) pp. 1655-1661 (1993).
Xu, et al., "Iterative Algebraic Reconstruction Algorithms for Emission Computed Tomogrpahy: A Unified Framework and its Application to Position Emission Tomography," Medical Physics--vol. 20(6), pp. 1675-1684 (1993).
Johnson, et al., "A Computer Model for the Study of Electrical Current Flow in the Human Thorax," Computers in Biology and Medicine--vol. 22(5), pp. 305-323 (1992).
Brandenburger, et al., "Effects of Anisotropy in the Ultrasonic Attenuation of Tissue on Computed Tomography," Ultrasonic Imaging--vol. 3(2), pp. 113-143 (1981).
Kudo, et al., "Feasible Cone Beam Scanning Methods For Exact Reconstruction in Three-Dimensional Tomography", Optical Society of America--vol. 7, No. 12, pp. 2169-2183 (Dec. 1990).
Feldkamp, et al., "Practical Cone-Beam Algorithm", Optical Society of America--vol. 1, No. 6, pp. 612-619 (Jun. 1984).
Kudo, et al., "Feasible Cone Beam Scanning Methods For Exact 3-D Tomographic Image Reconstruction", Prac. Topical Meeting OSA, Signal Recovery and Synthesis Cape Code, USA--pp. 174-177 (Jun. 1989).
Geman, et al., "A Comprehensive Statistical Model for Single-Photon Emission Tomography", Markow Random Fields--pp. 93-130 (1993).
Luo, et al., "Local Gemometry Variable Conductance Diffusion for Post-Reconstruction Filtering", IEEE Transactions on Nuclear Science--vol. 41, No. 6, pp. 2800-2806 (Dec. 1994).
Medoff, et al., "Iterative Convolution Backprojection Algorithms for Image Reconstruction from Limited Data", Optical Society of America--vol. 73, No. 11, pp. 1493-1500 (Nov. 1983).
Herman, et al., "Mathematical Methods in Tomography", Proceedings of a Conference Held in Oberwolfach, Germany--pp. 66-97 (Jun. 1990).
Payot, et al., "An Adaptive and Constrained Model for 3D X-Ray Vascular Reconstruction" Publication reference and date are not indicated.
Perona, et al., "Scale-Space and Edge Detection using Anisotropic Diffusion", IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 12, No. 7, pp. 629-639 (Jul. 1990).
Sauer, et al., "Bayesian Estimation of 3-D Objects from Few Radiographs", IEEE Transactions on Nuclear Science, vol. 41, No. 5, pp. 1780-1790 (Oct. 1994).
Smith, "Image Reconstruction from Cone-Beam Projections: Necessary and Sufficient Conditions and Reconstruction Methods", IEEE Transactions on Medical Imaging--vol. MI-4, No. 1, pp. 14-24 (Mar. 1985).
Guillemaud Regis
Payot Etienne
Preteux Francoise
Trousset Yves
Chang Jon
Dastouri Mehrdad
LandOfFree
Process for the reconstruction of a 3D image with contrast and r does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Process for the reconstruction of a 3D image with contrast and r, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process for the reconstruction of a 3D image with contrast and r will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-888704