*** 9661717 Hu This Small Business Innovation Research Phase I project will study the problem of reconstructing a three dimensional object from its projections acquired with a large area detector and cone-shaped x-ray beams. This problem is fundamental to the volumetric computerized tomography (CT) scanner development and has not yet been successfully solved. Based on the concept of function decomposition developed by the Principal Investigator, a "hybrid-type" reconstruction algorithm for the circle-and-line scanning orbit, consisting of a circular scan and a linear scan has been explored. In this project, the remaining problems associated with this algorithm will be studied. Specifically, the numerical accuracy and computational efficiency of this algorithm will be further improved by selecting the optimal theoretical formulation and numerical implementation. Furthermore, the algorithm will be extended to handle the object that is taller than the detector. This situation routinely occurs in practice. It will be demonstrated that this hybrid-type algorithm will combine the strengths and eliminate the weakness of existing algorithms and therefore could become the preferable algorithm for cone beam scanner development The successful development of this algorithm will accelerate the commercialization of the fast volumetric CT scanner. With drastic improvement in throughput and volume scanning capability, application of this technology will generate fundamental impacts in the fields of medicine, manufacturing industry, material science and engineering. ***
