Click here to


Are you sure ?

Yes, do it No, cancel

A Computational Model of Electronic Portal Imaging Device for Deterministic Particle Transport Based On Its Material Properties

I Yeo1*, J Yoon2 , N Joyce1 , A Gray1 , (1) University of Rochester, Rochester, NY, (2) Medical School of East Carolina Univ, Winterville, NC


(Sunday, 7/14/2019)  

Room: ePoster Forums

Purpose: To develop a computational model of an electronic portal imaging device (EPID) for Acuros in Eclipse treatment planning system using elemental compositions and densities of its structures.

Methods: The current Eclipse system in our institution is equipped with cross-section data of materials that are close to the structural materials of aSi-based EPID (Varian) in effective atomic numbers. They include Ti-Alloy, Stainless Steel, bone, water, and air. One of these materials was assigned to each structure of EPID; the density of the modeled structure was scaled based on the difference in effective atomic numbers between the assigned material and the actual composition of the structure. The thickness of each structure was rounded up/down to the nearest multiple of a whole millimeter, accounting for the smallest grid of Accuros. The scaled density was further adapted linearly to the determined thickness of each structure. In this way, the entire structure of EPID was modeled. Having placed the model (i.e. Phosphor) at 100cm SSD, we have calculated doses delivered to the phosphor from 6MV beams with the field sizes of 5x5, 10x10, and 20x20cm2. We have similarly irradiated EPID.

Results: The model showed its output factor (measurement/calculation) variation of 3.7 and -4.6% for 5x5 and 20x20cm2, respectively, relative to the output at 10x10cm2, smaller than that of the PD, 9.1 and -8.0%, respectively. As the model showed greater doses than measured doses at off-axis distances for all field sizes, when pixel-to-pixel calibration between measurements and calculations is employed using the largest field, its calculation will show better agreements with measurements.

Conclusion: The model, that allows deterministic particle transport, performed better than the PD, while consuming greater calculation time. It will be advantageous in cases when pre-treatment check fails. Its performance will be better evaluated in comparison to a Monte Carlo model of EPID.


Not Applicable / None Entered.


Not Applicable / None Entered.

Contact Email