Independent verification of monitor unit calculation for radiation treatment planning system

Chin J Cancer. 2010 Feb;29(2):217-22. doi: 10.5732/cjc.009.10463.

Abstract

Background and objective: To ensure the accuracy of dose calculation for radiation treatment plans is an important part of quality assurance (QA) procedures for radiotherapy. This study evaluated the Monitor Units (MU) calculation accuracy of a third-party QA software and a 3-dimensional treatment planning system (3D TPS), to investigate the feasibility and reliability of independent verification for radiation treatment planning.

Methods: Test plans in a homogenous phantom were designed with 3-D TPS, according to the International Atomic Energy Agency (IAEA) Technical Report No. 430, including open, blocked, wedge, and multileaf collimator (MLC) fields. Test plans were delivered and measured in the phantom. The delivered doses were input to the QA software and the independent calculated MUs were compared with delivery. All test plans were verified with independent calculation and phantom measurements separately, and the differences of the two kinds of verification were then compared.

Results: The deviation of the independent calculation to the measurements was (0.1 +/- 0.9)%, the biggest difference fell onto the plans that used block and wedge fields (2.0%). The mean MU difference between the TPS and the QA software was (0.6 +/- 1.0)%, ranging from -0.8% to 2.8%. The deviation in dose of the TPS calculation compared to the measurements was (-0.2 +/- 1.7)%, ranging from -3.9% to 2.9%.

Conclusions: MU accuracy of the third-party QA software is clinically acceptable. Similar results were achieved with the independent calculations and the phantom measurements for all test plans. The tested independent calculation software can be used as an efficient tool for TPS plan verification.

MeSH terms

  • Algorithms
  • Phantoms, Imaging
  • Quality Control
  • Radiometry / methods*
  • Radiotherapy Dosage
  • Radiotherapy Planning, Computer-Assisted / methods*
  • Software*