Room: AAPM ePoster Library
Purpose:
The work aims to determine the spectral response of photon
counting detectors (PCDs) from transmission measurements of a phantom of
known dimension and composition. Previously, this methodology has been
developed for low-flux transmission data, where the non-linear intensity
response of the PCD is not a factor. We extend the methodology
to measurements with higher photon flux.
Methods:
Simultaneous spectrum and non-linear intensity response calibration is performed
using transmission measurements of a step-wedge phantom with 25 known combinations
of Aluminum and PMMA thicknesses. The transmission measurements are obtained with
a DxRay PCD with 4 energy windowed transmission measurements. The X-ray source is
operated at an intensity so that the un-attenuated beam has a flux of 54% of the
maximum PCD count rate of 1,000,000 photons per second. The PCD spectral sensitivity
is arrived at by making a parametric model including the product of an initial source
spectrum estimate, a PCD sensitivity model, and an exponential of a 10th-degree polynomial
to allow the spectral sensitivity to adjust to the true response. Additionally, the ideal
transmitted intensity is taken as an input to a cubic polynomial to account for non-linear
effects such as detector pulse pile-up. The coefficients of the spectral exp-polynomial
and the intensity cubic polynomial are determined by fitting the model to the
Aluminum/PMMA step-wedge transmission data.
Results:
The PCD calibration arrived at from the step-wedge transmission measurements is tested
by predicting transmission for three independent materials: Teflon, Delrin, and Neoprene.
The results for testing the calibration model with and without the non-linear intensity fitting
are shown in the supplemental document.
Conclusion:
The additional non-linear intensity modeling greatly improves the PCD calibration for
high photon flux.
Not Applicable / None Entered.
Not Applicable / None Entered.