Analysis of spectral blur effects in x-ray scatter imaging
Medical Physics , Volume 26 - Issue 9 p. 1811- 1816
Previous analysis in our research program investigating the potential use of scattered photons for medical x-ray imaging has been for monoenergetic beams. In practice, polyenergetic beams are almost always used due to their higher photon fluence rate. The effects of beam polychromaticity on x-ray scatter imaging are determined with the aid of our semianalytic model that images a target object against a background material of the same dimensions when both are situated within a water phantom. Our analysis involves four different photon beams with constant incident energy fluence: (1) a monoenergetic beam with photon energy E0, (2) a dual peak beam with two separate monoenergetic peaks of energies E1 and E2, (3) a clinical x-ray beam, and (4) a rectangular beam with uniform energy fluence between energies E(min) and E(max). A comparison between the polyenergetic spectra is accomplished by matching the centroids and standard deviations of the dual peak and rectangular spectra to those of the clinical x-ray spectrum. For the task of imaging liver versus fat structures 1 cm thick in a 25-cm-diam spherical water phantom with the scattered photons between 2°and 12°, the predicted signal-to-noise ratio (SNR) obtained with a 100 kV beam is 87.5% of the SNR acquired with the optimum monoenergetic beam (SNR(opt). The SNR for the corresponding dual peak beam is 84.4% of SNR(opt) and for the rectangular beam is 86.3%. Our analysis shows that monoenergetic x-ray beams are not necessary for x-ray scatter imaging.