gms | German Medical Science

48th Meeting of the Particle Therapy Co-Operative Group

Particle Therapy Co-Operative Group (PTCOG)

28.09. - 03.10.2009, Heidelberg

A novel framework for beam angle optimization in intensity modulated hadron therapy

Meeting Abstract

Search Medline for

  • M. Bangert - Medical Physics, German Cancer Research Center, Heidelberg
  • U. Oelfke - Medical Physics, German Cancer Research Center, Heidelberg

PTCOG 48. Meeting of the Particle Therapy Co-Operative Group. Heidelberg, 28.09.-03.10.2009. Düsseldorf: German Medical Science GMS Publishing House; 2009. Doc09ptcog015

DOI: 10.3205/09ptcog015, URN: urn:nbn:de:0183-09ptcog0158

Published: September 24, 2009

© 2009 Bangert et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.



Background: To improve the automated treatment planning process in intensity modulated hadron therapy by integration of beam angle optimization (BAO). Besides beam angle selection based on dosimetric plan quality, the criterion of plan robustness with respect to range uncertainties is also considered.

Material and methods: A novel approach to BAO is introduced. Beneficial combinations of treatment beams are established prior to fluence optimization. The beam selection is based on a patient specific score which comprises a matrix of weighted ratios of dose delivered to the tumor to dose delivered to normal tissue and organs at risk for every target voxel and potential irradiation angle. The matrix of objectives is searched by an in-house developed genetic algorithm to identify optimal ensembles of treatment beams. It is possible to account for potential range uncertainties based on the consideration of a worst case score.

The general performance of the automated BAO framework is illustrated by a treatment plan comparison for an intra-cranial lesion. The robustness of the resulting solutions is illustrated by a treatment plan comparison for a prostate lesion.

Results: The BAO implemented in MatLab takes about 45 minutes on a state of the art work station. The main computational burden arises from the score calculation which scales linearly with the size of the target volume and the number of candidate beam directions. Typically, scores for ~5000 target voxels and ~1000 candidate beam directions are computed.

For the intra-cranial lesion, the dose to organs at risk is reduced applying the same number of optimized beams or less. At the same time, the optimized beam angle sets guarantee superior target coverage. DVHs are shown in Figure 1 [Fig. 1]. In contrast to the beam angles manually chosen by a radiation oncologist, the automated BAO yields an asymmetric beam configuration clustered around the main axis of the target volume.

For the prostate lesion, the optimized beam angle set differs by less than 20 degrees from the standard beam configuration with two opposed beams, but as depicted in Figure 2 [Fig. 2], the adjustments result in a treatment plan which is less sensitive to potential range uncertainties.

Conclusion: The selection of an adequate set of treatment beams is a critical factor for the overall quality of a hadron therapy treatment plan. The presented framework for BAO facilitates this process and yields improved and robust treatment plans for first clinical examples.

Acknowledgment: We thank Jule Daartz at FHBPTC at Massachusetts General Hospital for providing patient data and corresponding clinical orientations of proton beams.