Sorriaux, Jefferson
[UCL]
Rossomme, Séverine
[UCL]
Labarbe, Rudi
[IBA]
Bertrand, Damien
[IBA]
Salvat, Francesc
[Universitat de Barcelona]
Vynckier, Stefaan
[UCL]
Sterpin, Edmond
[UCL]
Introduction The main advantage of ion beam therapy over conventional therapy is an improved dose conformation and a significant lower dose to healthy tissues and organs at risk [1]. To take advantage of hadrons properties, treatment has to be delivered as it is planned with optimal accuracy. Pencil beam scanning (PBS) delivery technique is nowadays the most advanced technique for hadron beam therapy and needs proper quality assurances processes. For many years, Monte Carlo (MC) codes have been used to benchmark treatment planning systems (TPS) in conventional radiotherapy. Using a general purpose MC code, like GEANT4, it is possible to evaluate modalities such as those using combinations of different particles [2]. A pencil beam scanning model for proton treatment plan simulation using GEANT4/GATE platform has been validated in a water phantom for IBA PBS dedicated system [3]. In this work, we present preliminary results on validation in heterogeneous phantom using GEANT4/GATE model. Moreover simulations with a new MC code, called PENELOPE-proton (F. Salvat), were performed. Materials and methods 1. Measurements in heterogeneous phantom The pencil beam MC model was validated in a water phantom, reproducing pristine Bragg peaks for all tested energies (from 100 to 226.7 MeV) with accurcay of order of 0.7mm (range) and 0.2 mm (spot size). A spread-out Bragg peak with 10 cm modulation was reproduced with 0.8 mm range accuracy and a maximum dose difference of less than 2% point-to-point [3]. Measurements and simulation will be performed in similar beam delivery configurations in a heterogeneous phantom made of slabs collection with different densities and compositions (Fig 1.). Reference pristine Bragg peaks will be measured using plane parallel chamber in both configurations. Dose will be normalized to the number of protons via Faraday-cup measurements. Shifts for measurements are scheduled in January the Essen proton therapy center, Germany, with the IBA PBS dedicated nozzle. 2. Simulations We used the GEANT4-based toolkit GATE for active scanning beam simulations. The model proposes a generic method to generate scanned ion beam delivery system without simulation of the nozzle and based exclusively on beam data library measurements required for TPS [3]. We have also used an extension of PENELOPE to protons. PENELOPE-proton results were compared to data obtained with GATE. Moreover FLUKA MC code will be used as a third check. Results Differences of order of 4-5% were observed between the inelastic nuclear models G4BinaryCascade and G4PreCompound at entrance and in the plateau region. From previous work [3], binary cascade was found to be the best choice to match pristine Bragg peaks measurements but this should be confirmed in heterogeneous medium. PENELOPE code is still in programming development but it is worth emphasizing on the fact that a very simple algorithm is implemented in PENELOPE-proton to account for inelastic nuclear reactions with agreement to other codes comparable to Figure 2 whatever the energy and the phantom configuration chosen. For the PENELOPE-proton algorithm in its present configuration, with detailed simulation of electromagnetic interactions, time calculation for Pristine Bragg Peak simulation is reasonably of the order of the minute to reproduce full purpose MC results like GEANT4 ones. When accuracy of the code is ensured, simplifications in the electromagnetic processes will be performed to accelerate the simulation. Conclusion GATE was configured to deliver accurate IBA-PBS treatments. A final comparison with measurements in heterogeneous media, with a subsequent selection of the most appropriate nuclear models, is in progress. The PENELOPE-proton code delivered promising results, despite the simple description of nuclear inelastic interactions. Results offer perspectives for a future MC-based TPS involving the minimal physical complexity and maximum simulation speed to perform accurate dose distributions in patients.


Bibliographic reference |
Sorriaux, Jefferson ; Rossomme, Séverine ; Labarbe, Rudi ; Bertrand, Damien ; Salvat, Francesc ; et. al. Validation of a Monte Carlo pencil beam scanning model for proton in heterogeneous phantom using GEANT4/GATE and simulations with PENELOPE.Belgian Hospital Physicist Association Symposium 2012 (Brussels, du 03/02/2012 au 04/02/2012). |
Permanent URL |
http://hdl.handle.net/2078.1/128130 |