Comparison of the image lag for an aS500 EPID in direct and standard clinical configurations

  • Mrs Mahsheed Sabet, University of Newcastle, Australia
  • Mr Pejman Rowshan Farzad, University of Newcastle, Australia
  • Dr Brian King, University of Newcastle, Australia
  • Prof Frederick Menk, University of Newcastle, Australia
  • A/Prof Peter Greer, Calvary Mater Hospital Newcastle, and The University of Newcastle, Australia

Objective: The purpose of this study is to investigate the image lag for an a-Si EPID in both standard clinical and direct detection configurations, to improve EPID dosimetry measurements.
Methods: A Varian aS500 EPID was modified to direct configuration by replacing the layers sitting above the detector with a 30mm solid water buildup. A black sheet of paper was left on the detector and the experiments were performed in darkness. The image lag was investigated by:
- Taking continuous open field images, starting before beam-on and terminating when the shadow of the radiation field was not visible to the eye.
- Taking successive 5x5 followed by 10x10 images using high and low MUs respectively, with a separate single low MU image as reference. Comparison of the 10x10 fields with the reference could reveal the image lag.
- Taking an image in integrated mode followed by several dark field images in continuous mode.
- Taking three successive images in the integrated mode and investigating the increase in EPID signal.
The same steps were followed for a standard aS500 EPID.
Results: The experiments showed that the memory effect decreases to 1% of maximum signal within 2 seconds for both EPID configurations in 6 and 18 MV beams.
Conclusion: This study shows that the effect of image lag does not last long and would not affect IMRT dosimetry results, except for the last frames. The solution would be to continue imaging for 2 seconds after the beam is turned off.