Direct measurement of mammographic X-ray spectra with a digital CdTe detection system

Abstract

In this work we present a detection system, based on a CdTe detector and an innovative digital pulse processing (DPP) system, for high-rate X-ray spectroscopy in mammography (1-30 keV). The DPP system performs a height and shape analysis of the detector pulses, sampled and digitized by a 14-bit, 100 MHz ADC. We show the results of the characterization of the detection system both at low and high photon counting rates by using monoenergetic X-ray sources and a nonclinical X-ray tube. The detection system exhibits excellent performance up to 830 kcps with an energy resolution of 4.5% FWHM at 22.1 keV. Direct measurements of clinical molybdenum X-ray spectra were carried out by using a pinhole collimator and a custom alignment device. A comparison with the attenuation curves and the half value layer values, obtained from the measured and simulated spectra, from an ionization chamber and from a solid state dosimeter, also shows the accuracy of the measurements. These results make the proposed detection system a very attractive tool for both laboratory research, calibration of dosimeters and advanced quality controls in mammography.

Keywords: CdTe detectors; X-ray spectroscopy; digital pulse processing; high photon counting rate; mammography.

Figure 1.

Experimental set-up of clinical measurements of mammographic X-ray spectra with the digital system.

Figure 2.

(a) Throughput of the system: the measured photon counting rates from “fast” and “slow” channels (red and blue points, respectively) versus the true input counting rate. The red and blue lines are the paralyzable dead time model functions for the “fast” (dead time of 60 ns) and “slow” (dead time of 6.14 μs) channels, respectively; (b) The 22.1 keV photopeak centroid and energy resolution (FWHM) at 22.1 keV (percent deviation from the values at 200 cps).

Figure 3.

Measured ¹⁰⁹Cd spectra at 200 cps with no correction, at 830 kcps with no correction and at 830 kcps after PSD. The counts were normalized to the total number of detected events.

Figure 4.

Measured Ag- and W-target X-ray spectra (32 kV) with no correction and after PSD. The counts were normalized to the total number of detected events.

Figure 5.

Calculated tube voltage values vs. photon counting rate (W anode). Relative errors less than 0.22%, with a confidence level of 68%.

Figure 6.

The Mo-target X-ray spectra measured with the digital system under clinical conditions (28 kV and 30 kV, 20 mAs). The counts were normalized to the total number of detected events.

Figure 7.

Attenuation curves obtained from measured and simulated spectra and from direct exposure measurements (ionization chamber and solid state detector). The tube settings were: 28 kV and 20 mAs.