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

Tissue segmentation issues in Monte Carlo treatment planning for proton radiotherapy

Meeting Abstract

Suche in Medline nach

  • M. Bazalova - Stanford University, Stanford, USA
  • F. Verhaegen - Maastro Clinic, Maastricht, The Netherlands

PTCOG 48. Meeting of the Particle Therapy Co-Operative Group. Heidelberg, 28.09.-03.10.2009. Düsseldorf: German Medical Science GMS Publishing House; 2009. Doc09ptcog017

doi: 10.3205/09ptcog017, urn:nbn:de:0183-09ptcog0170

Veröffentlicht: 24. September 2009

© 2009 Bazalova et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen ( Er darf vervielfältigt, verbreitet und öffentlich zugänglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.



Background: Radiotherapy treatment planning is best done using the Monte Carlo (MC) method. Whereas MC dose calculations for conventional photon and electron radiotherapy have been studied extensively, proton beam MC dose calculations have only recently received attention. In this abstract, the utility of dual-energy CT (DECT) imaging for improved tissue segmentation (I) and the importance of metal streaking artifact reduction (II) for proton therapy MC treatment planning are studied.

Methods and materials: I. CT images of a solid water phantom with 9 tissue-equivalent inserts were acquired at two energies and segmented into materials and mass densities using the conventional single-energy CT method based on differences in relative electron densities and a more accurate DECT material extraction method, that makes use of the effective atomic number of each voxel. MC dose calculations for a broad 200 MeV proton beam were performed in the exact known geometry and in the single-energy and DECT geometries. II. CT images of a prostate patient with bilateral hip replacement were corrected using an algorithm based on interpolation of sinogram data. A 147 MeV proton beam treatment plan was created using MC on the basis of a water-only geometry, eleven PMMA slabs were used for beam energy modulation. MC dose calculations were subsequently performed for patient geometries based on the original CT images with severe streaking artifacts and based on the artifact corrected images. All dose distributions were calculated by the MCNPX code using the *F8:H,P,E energy deposition tally.

Results: I. Materials of three soft-bone tissue-equivalent inserts were incorrectly assigned using the conventional approach. Despite this, the dose calculation errors were below 2% in all miss-assigned media. Due to a ~0.05 g/cm3 inaccuracy in mass density assignment, the position of the Bragg peak in both the single-energy CT and DECT geometry was shifted by 0.7 cm. II. The patient dose calculations using CT images with streaking artifacts showed large statistical errors in the artifact corrupted prostate voxels that were incorrectly assigned to air. More importantly, due to the apparent air in the prostate, the 20% and 30% isodose lines extended by up to 1.5 cm into the healthy tissue in the artifact-corrupted geometry. The dose distribution using the artifact corrected CT images did not show this shift in isodose lines and predicted the dose distribution more accurately.

Conclusions: This work indicates that density assignment is more important than correct tissue segmentation for proton beam MC treatment planning. Therefore, DECT imaging for high-energy proton radiotherapy might only have a small benefit compared to the conventional tissue segmentation using single-energy CT images. In order to avoid an apparent shift in isodose lines into the healthy tissue, a metal artifact correction is necessary for prostate patients with hip prostheses receiving proton radiotherapy.