M Myronakis¹*, M Shi^1,2 , M Lehmann³ , P Huber³ , R Fueglistaller³ , M Jacobson¹ , P Baturin⁴ , D Ferguson¹ , I Valencia Lozano¹ , T Harris¹ , C Williams¹ , D Morf³ , R Berbeco¹ , (1) Brigham and Women's Hospital & Dana Farber Cancer Institute, Harvard Medical School, Boston, MA, (2) University of Massachusetts Lowell, Lowell, MA, (3) Varian Medical Systems, Baden, ZH, (4) Varian Medical Systems, Palo Alto, CA

• 
  M Myronakis

(Sunday, 7/14/2019) 1:00 PM - 2:00 PM

Room: 225BCD

Purpose: Development of novel flat-panel detector designs is limited by the long simulation times required for Monte Carlo methods. The main bottleneck is particle tracking within the detector. We have developed a tool to provide a major decrease in simulation time by pre-computing these interactions.

Methods: For the reduced particle tracking method, interactions within a novel flat-panel detector are precomputed for the relevant incident photon energy range. Then, during simulation, a GATE plugin takes the place of the detector model. For every incident photon, the detection by each layer of a multi-layer imager is determined based on a random number generator. If detected, the pre-computed optical spread function is added to the image. At the end, a complete image is returned to the user. We validated this method using a novel detector prototype based on four layers of standard Cu/GOS/a-Si imagers. Experimental data was acquired with a clinical linear accelerator at 6 MV, 6 MV-FFF and 2.5 MV. 2D slanted-slit images were used to calculate MTF and 2D projection images of a phantom were acquired for CBCT analysis. Measured images were compared to those simulated with the reduced particle tracking technique.

Results: Comparison between measured and simulated MTF values demonstrated very good agreement (RMSE =0.017). RMSE between simulated and measured HU values in CBCT volumes was 74 HU. An estimated time reduction on the order of 300 was found for the generation of the projection images of the phantom using the reduced particle tracking technique technique compared with conventional Monte Carlo simulations.

Conclusion: The reduced particle tracking technique was validated against MTF and HU measurements for a novel multi-layer imager architecture. This new method provides a major improvement in simulation time and will enable accelerated optimization of novel imager designs.

Funding Support, Disclosures, and Conflict of Interest: The project described was supported, in part, by Award Number R01CA188446 from the National Cancer Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute or the National Institutes of Health.

Keywords

Not Applicable / None Entered.

Taxonomy

Not Applicable / None Entered.

Contact Email