Encrypted login | home

Program Information

To Explore the More Realistic Energy Responses of the In-Depth Photon Counting Detectors

no image available
Y Yao

Y Yao*, N Pelc , Department of Bioengineering and Radiology, Stanford University, Stanford, CA


TU-F-18A-7 Tuesday 4:30PM - 6:00PM Room: 18A

Purpose: We study the effect of the secondary photon events on modeling the energy response functions of the In-depth photon counting X-ray detectors (PCXD) and the potential impact of the spectral distortion on material decompositions.

Methods: Square-shape wafers of three potential PCXD materials were constructed (5-by-20-by-30 mm^3 for Si, 4-by-20-by-5 mm^3 for GaAs and 4-by-20-by-3 mm^3 for CdTe), with pixel size of 5-by-4 mm^2 for Si and 5-by-5 mm^2 for GaAs and CdTe. The depth direction (z-direction) was segmented into 5 layers with exponentially increasing thicknesses of each layer. X-rays from 10keV to 120keV with 20000 photons per keV bin was simulated to characterize the energy response function of each PCXD using Geant4. Secondary photons events were recorded and we omitted the photons exiting the detector. The Energy Response Functions (ERFs) from the Monte Carlo (MC) simulations were compared with those from a semi-ideal model developed earlier.

Results: For Si, detection of secondary events in the center detector were minimal due to the long aspect ratio of the detector, which results in the agreement between the theoretical prediction and the MC simulation with and without the secondary photons. For CdTe, the secondary photons captured by the center pixel were important, leading to obvious disagreement between the analytical and the simulated ERF. After correction for secondary events, the two curves were more similar except for the escape peaks which are not correctly portrayed by the semi-ideal model. For GaAs, the behavior is in between Si and CdTe.

Conclusion: Given the complexity of the In-Depth PCXD’s geometry, the uniform semi-ideal model does not fully characterize the ERF at each layer. Therefore, more realistic models need to be explored for better modeling of the spectral distortion.

Contact Email: