Benchmarking of a new integrated 3D dosimetry system against Monte Carlo calculations and an established optical CT scanner.

Abstract

Background: Advanced radiation therapy techniques, including intensity-modulated radiation therapy (IMRT), stereotactic radiosurgery (SRS), adaptive therapy, and proton therapy, offer high precision in delivering radiation doses to tumors while minimizing exposure to surrounding healthy tissues. These sophisticated methods necessitate stringent quality assurance (QA) measures to ensure their accuracy and safety. Three-dimensional (3D) dosimetry systems have the potential to play an important role in this context for verifying dose distributions in a comprehensive manner but have not been widely implemented partially due to a lack of streamlined systems that include dosimeter, readout, and analysis.

Purpose: The ClearView radiochromic dosimeter, the Vista 16 Optical CT scanner, and the VistaAce analysis software have the potential as a fully integrated 3D dosimetry tool for commissioning and verifying complex radiotherapy treatment plans. We aim to benchmark this integrated 3D dosimetry system and investigate its clinical utility.

Methods: The performance of this system was benchmarked against an independent Monte Carlo dose calculation software, the Duke Large Field of View Optical CT Scanner (DLOS), and an open-source analysis software (3D Slicer v4.13). We measured two simple radiotherapy plans and a selection from the AAPM (American Association of Physicists in Medicine) Task Group 119 IMRT commissioning tests. Treatment plans were prepared within the Eclipse planning system (AAA v15.6.03) after which a Varian Truebeam linac was used to deliver the treatment plans. Vista 16 was used to reconstruct the measured 3D dose distribution which was compared to the dose distribution obtained from an independent Monte Carlo-based dose calculation algorithm, as well as the 3D dose distribution reconstructed using the well-established DLOS. Image registration, conversion from optical density to dose, and comparative analysis were done using the VistaAce software and validated against results obtained using 3D Slicer for a subset of tests.

Results: ClearView dosimeters exhibited a linear dose-response up to 60 Gy. For the 3-field benchmarking irradiation, the agreement (2%/2 mm 3D global gamma Index, 10% threshold) between ClearView/VistaAce versus the TPS and Monte Carlo was 97.8% and 98.8%, respectively. For the AAPM TG119 mock head and neck plan, the agreement (2%/2 mm) with the treatment planning system and Monte Carlo was 99.1% and 95.1%, respectively. For the TG119 mock prostate, the agreement was 99.7% and 98.9%, respectively. Agreement for the ClearView/ Vista 16 dose reconstruction was equivalent or superior to that of the ClearView/DLOS reconstruction for the benchmarking irradiations.

Conclusion: The ClearView/Vista 3D dosimetry system demonstrated robust performance in measuring and verifying realistic clinical dose distributions, with good agreement with an independent Monte Carlo algorithm and equivalent or better agreement than DLOS. The system's integrated approach, combining dosimetry, scanning, and analysis, streamlines QA processes in advanced radiation therapy, potentially enhancing clinical practice by providing consistent and accurate dosimetric verification.