Encrypted login | home

Program Information

A Comparison of Multi-Detector CT Energy Spectrums for Use in Monte Carlo Simulations


G Miller

G Miller1*, S Ahmad2, C Ruan3, (1) Washington Hospital Center, Washington DC, (2) Oklahoma Univ. Health Science Ctr., Oklahoma City, OK, (3) University of Oklahoma Health Science Center, Oklahoma City, OK

SU-E-I-32 Sunday 3:00PM - 6:00PM Room: Exhibit Hall

Purpose: To compare two different methods of obtaining energy spectrums for use in Monte Carlo (MC) simulations of a multi-detector CT (MDCT) scanner.

Methods: The first set of CT energy spectrums were obtained by directly measuring the x-ray tube output using an Amptek CdTe detector with a collimator. The CT x-ray tube was fixed at the 12 o'clock position and the detector was placed on the patient table. The bowtie filters were removed from the path of the beam, and the tube current was set to 10 mA, the lowest possible setting, to minimize x-ray fluence. Data was recorded for tube potentials of 80, 100, 120, and 140 kVp. However, the 140 kVp potential saturated the detector. The second set of CT energy spectrums was derived by utilizing a MC simulation to pass a soft tungsten anode spectrum through a proprietary beam hardening material and recording the resulting spectrum. The initial tungsten anode spectrum was obtained using an implementation of Boone and Seibert's tungsten anode spectral model.

Results: The MDCT scanner uses a tungsten anode which theoretically results in a spectrum with characteristic x-rays of 59 keV and 67 keV. The second characteristic x-ray peak was clearly evident in all of the measured spectrums; however, the first characteristic x-ray peak was almost non-existent. The derived spectrums were similar to the measured spectrums in that most of the soft energy in the 30-50 keV range had been filtered out. However, both characteristic x-rays were present in the derived spectrums.

Conclusion: Different spectrums resulting from measured and derived methods suggest a machine specific spectrum should be acquired for accurate Monte Carlo patient dose evaluation. The attenuation in the measured spectrum indicates the presence of a fairly significant hardening filter which may not have been fully accounted for when developing the derived spectrums.

Contact Email: