Phase-contrast computed tomography (CT) have advantages of analyzing low Z objects such as polymer and soft tissue. Especially, X-ray grating interferometer CT is a practical method to obtain phase-contrast CT, but it has limited object size because of the limitation of the grating size. So, if the object is larger, the interior problem is occurred. It is known that there is no exact solution to solve this problem. In this study, we used machine learning to reduce the artifacts due to data truncation. We prepared the first input as a filtered backprojection (FBP) output, which is a classical image reconstruction method that has severe artifacts when data is truncated. And we also prepared the second input as geometrical information to clarify the region of interest (ROI). These networks were compared in two cases; a single input, two inputs. Visual results and quantitative results were used to compare image quality about various methods. Simulation results showed the better results than other methods. Our results show that machine learning is a promising technique to solve the CT challenges, may have many applications to all imaging fields.
Over the past two years, we have developed a series of algorithms for Grangeat-type half-scan based reconstruction of a short object. These algorithms allow high temporal resolution and high temporal consistence, and suppress Feldkamp-type reconstruction related artifacts. Therefore, this scheme is promising for dynamic and/or quantitative imaging. In this paper, we extend our work into a solution to the long object problem. Our approach takes a temporally non-optimized pre-reconstruction step to transform the long object problem into a short object problem. The detector area is analytically classified into desirable, corrupted, and useless areas. The cone-beam data in the corrupted area are then corrected by the forward projection of the pre-reconstruction, while the data in the useless area are set to zero. A generalized Feldkamp algorithm is chosen for the pre-reconstruction. After the correction, the Grangeat-type half-scan based reconstruction of a short object is performed along with several shadow zone interpolation techniques for the final reconstruction. Numerical simulation is conducted to compare the proposed algorithm with a half-scan Feldkamp algorithm.
In this paper, we perform numerical studies on Feldkamp-type and Katsevich-type algorithms for cone-beam reconstruction with a nonstandard spiral locus to develop an electron-beam micro-CT scanner. Numerical results are obtained using both the approximate and exact algorithms in terms of image quality. It is observed that the two algorithms produce similar quality if the cone angle is not large and/or there is no sharp density change along the <i>z</i>-direction. The Katsevich-type algorithm is generally preferred due to its nature of exactness.
Currently, various cone-beam CT scanners are under rapid development for major biomedical applications. Half-scan cone-beam image reconstruction algorithms are desirable, which assume only part of a scanning turn, and are advantageous in terms of temporal resolution. While the existing half-scan cone-beam algorithms are in the Feldkamp framework, we formulate a half-scan algorithm in the Grangeat framework for circular and helical trajectories. First, we modify the Grangeat formula in the circular half-scan case. With analytically defined boundaries, the Radon space is partitioned into shadow zone, singly and doubly sampled regions, respectively. A smooth weighting scheme is designed to compensate for data redundancy and inconsistency. The sampled regions are linearly interpolated into the shadow zone for a complete data set. Then, these concepts and formulas are extended to the helical half-scan case. Extensive numerical simulation studies are performed to verify the correctness and demonstrate the performance. Our Grangeat-type half-scan algorithms allow minimization of redundant data and optimization of temporal resolution, and outperform Feldkamp-type reconstruction in terms of image artifacts. These algorithms seem promising for quantitative and dynamic biomedical applications of cone-beam tomography.