No abstract

Purpose: The method of obtaining the primary x‐ray spectrum by the Compton scattering correction has been established for relatively high x‐ray tube voltages. However, the influence of Rayleigh scattering can not be neglected in the mammography energy range. Besides, accurate data of Rayleigh and Compton cross sections for the scatterer material are required for obtaining the primary x‐ray spectrum by correcting the scattered x‐ray spectrum in mammography. The purpose of this study is to investigate theoretically and experimentally the relative contributions of Rayleigh scattering for PMMA (polymethyl methacrylate) in the mammography energy range. Method and Materials: Fluorescent x‐rays in the energy range of 10 to 25 keV were generated by exciting four metal targets including molybdenum by synchrotron radiation. The fluorescent x‐rays were incident on a sphere of PMMA, and the scattered x‐rays were measured with a CdTe detector at scattering angles of 90, 120, 150, 165 degrees. The scattered fluorescent x‐ray spectrum was separated into the Rayleigh and Compton scattering peaks by using a curve fitting technique with two Gaussian functions. We compared the measured cross section data to those obtained from theoretical values. We also calculated the scattered fluorescent x‐ray spectra using the Monte Carlo simulation and compared them to the measured spectra. Results: The measured ratio of Rayleigh to total scattering cross section was about 30% at 10 keV. The maximum discrepancy of the measured and theoretically calculated values of the ratio was about 30% for the scattering angles of 90 to 165 degrees. However, at the angle of 120 degrees where the overall error might be minimum, the theoretical and experimental values agreed within 2%. Conclusion: Rayleigh and Compton scattering cross sections should be measured more precisely in order to improve the calculation of the primary mammography x‐ray spectrum from the scattered x‐ray spectrum.

Purpose: An analysis of the x‐ray spectrum is important for quality assurance (QA) and quality control (QC) of a radiographic system. In the case of mammography, the direct measurement of the primary x‐ray spectra under clinical conditions is very difficult and time‐consuming. The method of obtaining the primary x‐ray spectrum by the Compton scattering correction has been established for relatively high x‐ray tube voltages. However, the influence of Rayleigh scattering can not be neglected for mammography. We have developed a new method of reconstructing the primary x‐ray spectrum from the scattered x‐ray spectrum taking into account both Rayleigh and Compton scattering. Method and Materials: The primary x‐ray beam from a 28 kV mammography x‐ray unit was incident on a PMMA (polymethyl methacrylate) sphere scatterer with a diameter of 6 mm. The 90‐degree scattered x‐ray spectrum was measured by using a CdTe semiconductor detector. The measured x‐ray spectrum was separated into three energy regions, and the characteristic x‐ray peaks and bremsstrahlung x‐rays were fitted by Gaussian and quadratic functions. The energy shift corrections of the Compton and Rayleigh components were made at each photon energy channel by splitting the number of photons at each channel into the two scattering components according to the corresponding cross sections. The Monte Carlo simulation of the 90‐degree scattered x‐ray spectra were also performed using the EGS5 code. Results: The reconstructed spectrum agreed fairly well with a directly measured primary x‐ray spectrum. In the Monte Carlo simulation, the scattered x‐ray spectrum calculated for the incidence of the reconstructed x‐ray spectrum showed a very good agreement to the measured scattered x‐ray spectrum. Conclusions: The Rayleigh and Compton scattering correction method could be suitable for measuring the mammography x‐ray spectra under clinical conditions and useful for QA and QC of the mammography x‐ray units.

No abstract