Weijie Tao1,2,3, Yongjin Sung4, Sally Ji Who Kim5, Qiu Huang1,2, Grant T Gullberg3, Youngho Seo3, Michael Fuller6. 1. Department of Nuclear Medicine, Rui Jin Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China. 2. School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China. 3. Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, USA. 4. Department of Biomedical and Mechanical Engineering, University of Wisconsin-Milwaukee, Milwaukee, WI, USA. 5. Cardiovascular Research Center, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. 6. TF Instruments, Salinas, CA, USA.
Abstract
PURPOSE: In this work, we present tomographic simulations of a new hardware concept for X-ray phase-contrast interferometry wherein the phase gratings are replaced with an array of Fresnel biprisms, and Moiré fringe analysis is used instead of "phase stepping" popular with grating-based setups. METHODS: Projections of a phantom consisting of four layers of parallel carbon microfibers is simulated using wave optics representation of X-ray electromagnetic waves. Simulated projections of a phantom with preferential scatter perpendicular to the direction of the fibers are performed to analyze the extraction of small-angle scatter from dark-field projections for the following: (1) biprism interferometry using Moiré fringe analysis; (2) grating interferometry using phase stepping with eight grating steps; and (3) grating interferometry using Moiré fringe analysis. Dark-field projections are modeled as projections of voxel intensities represented by a fixed finite vector basis set of scattering directions. An iterative MLEM algorithm is used to reconstruct, from simulated projection data, the coefficients of a fixed set of seven basis vectors at each voxel representing the small-angle scatter distribution. RESULTS: Results of reconstructed vector coefficients are shown comparing the three simulated imaging configurations. The single-exposure Moiré fringe analysis shows not only an increase in noise because of one-eighth the number of projection samples but also is obtained with less dose and faster acquisition times. Furthermore, replacing grating interferometry with biprism interferometry provides better contrast-to-noise. CONCLUSION: The simulations demonstrate the feasibility of the reconstruction of dark-field data acquired with a biprism interferometry system. With the potential of higher fringe visibility, biprism interferometry with Moiré fringe analysis might provide equal or better image quality to that of phase stepping methods with less imaging time and lower dose.
PURPOSE: In this work, we present tomographic simulations of a new hardware concept for X-ray phase-contrast interferometry wherein the phase gratings are replaced with an array of Fresnel biprisms, and Moiré fringe analysis is used instead of "phase stepping" popular with grating-based setups. METHODS: Projections of a phantom consisting of four layers of parallel carbon microfibers is simulated using wave optics representation of X-ray electromagnetic waves. Simulated projections of a phantom with preferential scatter perpendicular to the direction of the fibers are performed to analyze the extraction of small-angle scatter from dark-field projections for the following: (1) biprism interferometry using Moiré fringe analysis; (2) grating interferometry using phase stepping with eight grating steps; and (3) grating interferometry using Moiré fringe analysis. Dark-field projections are modeled as projections of voxel intensities represented by a fixed finite vector basis set of scattering directions. An iterative MLEM algorithm is used to reconstruct, from simulated projection data, the coefficients of a fixed set of seven basis vectors at each voxel representing the small-angle scatter distribution. RESULTS: Results of reconstructed vector coefficients are shown comparing the three simulated imaging configurations. The single-exposure Moiré fringe analysis shows not only an increase in noise because of one-eighth the number of projection samples but also is obtained with less dose and faster acquisition times. Furthermore, replacing grating interferometry with biprism interferometry provides better contrast-to-noise. CONCLUSION: The simulations demonstrate the feasibility of the reconstruction of dark-field data acquired with a biprism interferometry system. With the potential of higher fringe visibility, biprism interferometry with Moiré fringe analysis might provide equal or better image quality to that of phase stepping methods with less imaging time and lower dose.
Authors: Arvidas B Cheryauka; James N Lee; Alexei A Samsonov; Michel Defrise; Grant T Gullberg Journal: Magn Reson Imaging Date: 2004-02 Impact factor: 2.546
Authors: A F Isakovic; A Stein; J B Warren; A R Sandy; S Narayanan; M Sprung; J M Ablett; D P Siddons; M Metzler; K Evans-Lutterodt Journal: J Synchrotron Radiat Date: 2010-05-18 Impact factor: 2.616
Authors: Timm Weitkamp; Ana Diaz; Christian David; Franz Pfeiffer; Marco Stampanoni; Peter Cloetens; Eric Ziegler Journal: Opt Express Date: 2005-08-08 Impact factor: 3.894