Encrypted login | home

Program Information

A Custom Web Application for a GPU-Based Monte Carlo IMRT/VMAT QA Tool

no image available
M Folkerts

M Folkerts*1,2, Y Graves1,2, Q Gautier2, G Kim2, X Jia2, S Jiang2, (1) Department of Physics, (2) Center for Advanced Radiotherapy Technologies, University of California, San Diego, La Jolla, CA

SU-E-T-193 Sunday 3:00PM - 6:00PM Room: Exhibit Hall

Purpose: To develop a user friendly web application for a GPU-based Monte Carlo (MC) 3D dosimetry quality assurance (QA) tool which runs in modern web browsers and interacts remotely with specialized hardware/software resources.
Methods: We developed a new QA web application based on existing technologies (HTML5, Python, and Django) to interface with a command line-based MC QA tool. It eliminates the need for users of this cutting-edge GPU software to purchase expensive/specialized hardware and use a command-line based environment. The result is a web app with a user-friendly interface enabling control of GPU accelerated software for MC-based QA. Users upload zipped DICOM RP, CT, and RD data and set the expected statistic uncertainty for the MC dose calculation. After the files are uploaded, the remote software is automatically launched to generate intermediate data necessary for the GPU-based MC dose calculation. The MC dose calculation is then executed on the remote GPU server and the resultant dose and original plan dose (from uploaded RD file) are displayed in the web GUI for the user to review. A 3D gamma index map and DVH curves are also displayed to the user. Finally, a PDF QA report that summarizes the results can be downloaded.
Results: We successfully developed a web app for a GPU-based QA tool that consists of fluence map generation, CT image processing, GPU based MC dose calculation, gamma index calculation, and DVH calculation. The result is an IMRT and VMAT QA tool that conducts an independent dose calculation for a given treatment plan. The computation time from data uploading to viewing results and downloading report is less than 2 min.
Conclusion: We developed a GPU-based MC QA tool with a user-friendly web-based interface. The high efficiency and accessibility greatly facilitate IMRT and VMAT QA.

Contact Email: