Click here to

Session: Imaging: Mammography and Tomosynthesis [Return to Session]

Monte-Carlo Simulation of the Detective Quantum Efficiency of Cadmium Telluride X-Ray Detectors for Photon-Counting Mammography

J Day*, J Tanguay, Ryerson University, Toronto, ONCA,


SU-IePD-TRACK 2-4 (Sunday, 7/25/2021) 3:00 PM - 3:30 PM [Eastern Time (GMT-4)]

Purpose: Photon counting offers the possibility of single-shot spectroscopic imaging and electronic noise rejection, but charge sharing will degrade image quality. We used Monte Carlo (MC) methods to simulate the detective quantum efficiency (DQE) of cadmium-telluride (CdTe) photon-counting x-ray detectors for breast imaging and compared results against published DQEs for energy-integrating systems that use amorphous selenium (a-Se) detectors.

Methods: We calibrated and validated a MC model of x-ray interaction and detection in CdTe photon-counting x-ray detectors against an XCounter Thor detector. Our model accounted for charge sharing, electronic noise and anticoincidence logic. We then optimized the detector thickness for a 28kVp Mo/Mo spectrum hardened by 3.9cm of lucite as a function of electronic noise level and pixel size with and without inter-pixel anticoincidence logic. The figure of merit used for optimization was the integral of the DQE. The optimized DQE was compared to published values of the DQE of energy-integrating a-Se systems for 85um pixels.

Results: Without anti-coincidence logic, the optimal DQE(0) was 0.72 at the optimal detector thickness of 300 um, which is comparable to the DQE of a-Se systems in clinical use. Anti-coincidence logic reduced the photon-counting DQE(0) by 14% due to an associated increase in electronic noise. Reducing the electronic noise level to ~1.5 keV in combination with anti-coincidence logic resulted in a DQE(0) of 0.84.

Conclusion: Increasing the DQE of photon-counting CdTe detectors relative to that of energy-integrating a-Se detectors will require the use of anti-coincidence logic to suppress charge sharing, and electronic noise levels to less than ~1.5keV, which is approximately half of that of the system used in this work. Future work will focus on contrast-enhanced spectral mammography.

Funding Support, Disclosures, and Conflict of Interest: Natural Sciences and Engineering Research Council of Canada Discovery Grants Program. Canadian Foundation for Innovation John R. Evans Leaders Fund



    Monte Carlo, Spectroscopic Imaging, Mammography


    IM- X-Ray: Monte Carlo modeling

    Contact Email