Three-Dimensional Reconstruction Algorithm Using Corn-beam X-ray

Research Project

Project/Area Number	60550292
Research Category	Grant-in-Aid for General Scientific Research (C)
Allocation Type	Single-year Grants
Research Field	計測・制御工学
Research Institution	Nagoya Institute of Technology
Principal Investigator	SUZUMURA Nobuo Nagoya Inst. of Tech. Facility of Eng., Professor, 工学部, 教授 (10023102)
Co-Investigator(Kenkyū-buntansha)	IWATA Akira Nagoya Inst. of Tech. Facility of Eng., Associate Professor, 工学部, 助教授 (10093098) ISHII Naohiro Nagoya Inst. of Tech. Facility of Eng., Professor, 工学部, 教授 (50004619)
Project Period (FY)	1985 – 1986
Project Status	Completed (Fiscal Year 1986)
Budget Amount *help	¥2,200,000 (Direct Cost: ¥2,200,000) Fiscal Year 1986: ¥800,000 (Direct Cost: ¥800,000) Fiscal Year 1985: ¥1,400,000 (Direct Cost: ¥1,400,000)
Keywords	CT / Three dimensional CT / 画像再構成
Research Abstract	X-ray computed tomography(X-ray CT) can provide the internal structure of the cross-section of the human-body; as the images of X-ray absorption coefficient. Though conventional CT provides only a transversal image of the body or organs of interest, three-dimensional(3-D) structures of organs are needed for clinical diagnosis and radiation therapy planning. Conventional technique to make 3-D image is to synthesize a series of 2-D images produced by sequential scans. However, the spatial resolution of the synthesized images with a respect to sagittal, and coronal direction are inferior to that of transversal direction. In this research we studied about the reconstructing algorithm for divergent X-ray CT, which can provide 3-D image with a rotational scan. But projection data generated by a single rotational scan of divergent X-ray do not give complete information of the object. This incompletion is the point to be special considered. We have investigated the filtered back-projection method, which provides the reconstructed images with high quality. As the result we found a new filter function. We call it modified <rho> filter, which consists of two parts; (1) A deblurred filter which is high-path filter with different characteristics in the direction of parallel and perpendicular to the rotational axis is used. (2) A space variant lowpass filter in the direction of rotational axis is used for the purpose of reducing the artifacts caused by incomplete projection data. The evaluation of the algorithm has been done by computer simulaiton. The results are following; (1) X-ray absorption coefficients have been accurately reconstructed, even using incomplete projection data. (2) The artifacts caused by incomplete projections have been reduced significantly without much compromise of spatial resolution in the direction of rotational axis.