Diapositive 1

❑

Use of a common data set as input for 7 commercial planning systems

❑

Test package

❑

Comparison: Percentage deviations of the local dose except points ourside the

penumbra or under blocs where the deviation was expressed relatively to the dose on

the central axis of the open beam.

❑

Confidence límit:

❑

Tolerance: depending on the region

Performance testing – Test package example

Appl ication of a test package in an intercomparison of the

photon dose calculation performance of treatment planning systems

used in a cl inical setting

Jack Venselaar

* , Hans Wel leweerd

Department of Radiotherapy, Dr B. Verbeeten Institute, P.O. Box 90120, 5000 LA Tilburg, The Netherlands

Department of Radiotherapy, University Medical Center, P.O. Box 8500, 3508 GA Utrecht, The Netherlands

Received 26 May 2000; received in revised form 12 December 2000; accepted 9 January 2001

Abstract

Background and purpose: T sting the performance of treatmen planning system by using the AAPM Task Group 23 t st package is a

useful approach, but has its limitations. To be able to include technical developments, such as the asymmetric collimator, it was decided to

remeasure the AAPM data set on more modern radiotherapy equipment, to extend the test geometries, and to evaluate the use of the new

package.

Materials a d m thods: A coherent set of beam data of 6, 10 and 18 MV photon beamswasmeasured on two modern linear accelerators.

These data served as input data in seven commercially available treatment planning systems, which were clinically in use in different

radiotherapy departments. Next, a test package was measured which included a missing tissue geometry and ®elds with asymmetrical

collimator setting, with and without a wedge.

Results: The absolute dose prediction from the different treatment planning systems in which themeasured beam datawere entered, was

compared for all test pointswith the resultsof direct measurements. Thecriteriaof acceptability wereexceeded by somesystems in casesof

irregular ®eld geometry and missing tissue geometry. The majority of the systems had dif®culties with accurate dose calculation for

asymmetrically wedged ®elds.

Conclusions: The application of the new test package did not introduce insuperable dif®culties and was highly appreciated by the

participating centres. Most systemsperformed reasonably well for themajority of thebeamgeometries, with theexception of asymmetrically

wedged beams. The extended test package is available for other users or user groups for the purpose of commissioning new treatment

Radiotherapy and Oncology 60 (2001) 203±213