Click here to


Are you sure ?

Yes, do it No, cancel

Implementation and Benchmarking of Volumetric Modulated Arc Therapy (VMAT) Modeling in the GPU-Based High-Performance Monte Carlo Code ARCHER

D Adam1*, T Liu2 , P Caracappa3 , X Xu2 , B Bednarz1 , (1) University of Wisconsin- Madison, Madison, WI, (2) Rensselaer Polytechnic Institute, Troy, NY, (3) Virtual Phantoms, Inc., Albany, NY


(Monday, 7/15/2019) 4:30 PM - 6:00 PM

Room: 301

Purpose: ARCHER is a new Monte Carlo-based radiation therapy dose engine that leverages both NVIDIA and AMD GPUs to significantly reduce computational time yet retain the dose calculation accuracy of the Monte Carlo radiation transport method. Previous versions of ARCHER were only capable of modeling step-and-shoot IMRT treatment deliveries. In this work we have implemented VMAT modeling in ARCHER and benchmarked VMAT simulations for a Varian TrueBeam STx 6X accelerator in ARCHER to VMAT simulations performed in the EGSnrc code package.

Methods: Validation of ARCHER was conducted by comparing against a well benchmarked EGSnrc model of a TrueBeam STx 6X. Multiple benchmarking scenarios were conducted including an MPPG5a MLC test pattern case and one clinical prostate VMAT patient plan. Dose difference maps and gamma analyses were conducted to establish the accuracy and consistency between the two Monte Carlo models. Timing studies were conducted on both NVIDIA and AMD GPUs to evaluate the computational speed of VMAT plans calculated using ARCHER.

Results: For the MPPG5a MLC test pattern, the 3%/3mm gamma pass rate between EGSnrc and ARCHER was 99.1%. For the VMAT prostate patient, the 3%/3mm gamma pass rate between EGSnrc and ARCHER was 99.0%. Timing studies conducted demonstrated that dose calculation of VMAT plans can be conducted in as little as1 minute, which is much faster than dose calculations conducted using EGSnrc.

Conclusion: A patient specific, VMAT capable, model of the Varian TrueBeam STx 6X has been validated in ARCHER by benchmarking against EGSnrc. With the addition of VMAT capabilities, this work demonstrates that Monte Carlo based dose engines like ARCHER have the potential for widespread clinical implementation.

Funding Support, Disclosures, and Conflict of Interest: Grant support was provided from the National Institute of Biomedical Imaging and Bioengineering (STTR 4R42EB019265)


EGS4, Monte Carlo, Treatment Planning


TH- External beam- photons: dose computation engines- Monte Carlo

Contact Email