Direct measurement of mammographic x-ray spectra using a CdZnTe detector

Our purpose is to directly measure mammographicx-ray spectra with collimators and a low-efficiency CdZnTe detector developed recently and to find out the best fit response function of CdZnTe detector to correct the measured spectra.Photonspectra (target Mo or Rh) produced by a mammographicx-ray unit at 25–32 kV and 240 mAs (=3 times of 80 mAs) and transmitted through 0.03 mm Mo or 0.025 mm Rh filter and object (0.1 mm Al to 0.8 mm Al phantoms) have been analyzed. Since detectedspectra were distorted by the response of CdZnTe detector and did not present the true photonspectra, the correction was applied by the stripping procedure. The response function of detector used in this procedure has been determined by the evaluation of interactions (K-escape, coherent scattering, and Compton scattering processes) and incomplete charge collection calculated using the Monte Carlo method. We have used K α1 , K α2 , K β1 , K β2 radiations of Cd, Zn, and Te, respectively and have used the weight function for the incomplete charge collection and have considered Compton scattering. The Monte Carlo simulations were continued by changing the important factors (mean path length of hole λ h , dead layer of the CZT crystal and weight factor W q ) of incomplete charge collection until the best fit response function was found out. Corrected photonspectra were compared with the mammographicx-rayspectral data of Bureau of Radiological Health (BRH) measured using a Ge detector. Attenuation curves of aluminum for 25–32 kV were calculated from the corrected photonspectra and compared with the attenuation curves measured using an ionization chamber. These results obtained using the CdZnTe detector agreed with the mammographicx-rayspectral data of BRH and attenuation curves obtained by the ionization chamber.