MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

Implementation of a Novel Cloud-Based Monte Carlo Application for Delivery Performance Monitoring and Secondary Dose Calculation for Tomotherapy: A Multiple-Treatment-Site Study

L Sensoy1*, Q Chen2, Y Rong3, S Benedict4, (1) UC Davis Medical Center, Sacramento, CA, (2) University of Kentucky, Lexington, KY, (3) University of California-Davis, Sacramento, CA, (4) UC Davis Cancer Center, Sacramento, CA

Presentations

(Sunday, 7/12/2020)   [Eastern Time (GMT-4)]

Room: AAPM ePoster Library

Purpose:
Secondary calculations and plan checks have been a challenge for Tomotherapy. In this study we evaluate and validate a new cloud-based comprehensive Monte Carlo (MC) application for delivery performance monitoring and dose verification calculations of Tomotherapy treatment plans for four different treatment sites.

Methods:
A cloud-based MC code package for Tomotherapy was used to provide delivery performance monitoring and secondary dose calculations. All patient data were anonymized prior to transmission for calculations on the external cloud-based server. MC calculation run-time is no more than 10 minutes per plan with an uncertainty parameter set to 0.03. The MC package was evaluated on four treatment sites: Head and Neck, Brain, Pelvis, and Prostate. For each site fifteen plans were evaluated by MC-based dose value histogram (DVH) verification. In addition, using the post-treatment machine log files and input from the treatment planning system (TPS), a second MC calculation (MCLogQA) utilizing the sinogram acquired by the exit detector after the first fraction of the treatment, was used for delivery performance monitoring. MCLogQA was then compared to TPS and MC for agreement. These comparisons and evaluation are not available on any commercial system.

Results:
The MCLogQA results agree well with both TPS and MC, generally within 3% and 1%, respectively, except for low dose regions outside of the field. The D5, D50, and D95 dose calculations from MC and MCLogQA were all within acceptable agreement to TPS, for all treatment sites. In most of the cases, MCLogQA was either in complete agreement with or predicting slightly less dose compared to MC.

Conclusion:
The studied cloud-based MC tool is a fast, inexpensive ($0.10/plan), semi-automated alternative to widely practiced clinical standard using phantoms and films for patient-specific QA. The MC package also provides an essential tool, MCLogQA, which is highly instrumental in identifying treatment delivery issues.

Keywords

Not Applicable / None Entered.

Taxonomy

Not Applicable / None Entered.

Contact Email