Related Experiment Video
Updated: Dec 13, 2025
Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System
Published on: April 11, 2018
Automated Monte-Carlo re-calculation of proton therapy plans using Geant4/Gate: implementation and comparison to
Adam H Aitkenhead1,2, Peter Sitch1, Jenny C Richardson1,2
1Christie Medical Physics and Engineering, The Christie NHS Foundation Trust, Manchester, UK.
Objectives:
Software re-calculation of proton pencil beam scanning plans provides a method of verifying treatment planning system (TPS) dose calculations prior to patient treatment. This study describes the implementation of AutoMC, a Geant4 v10.3.3/Gate v8.1 (Gate-RTion v1.0)-based Monte-Carlo (MC) system for automated plan re-calculation, and presents verification results for 153 patients (730 fields) planned within year one of the proton service at The Christie NHS Foundation Trust.
Methods:
A MC beam model for a Varian ProBeam delivery system with four range-shifter options (none, 2 cm, 3 cm, 5 cm) was derived from beam commissioning data and implemented in AutoMC. MC and TPS (Varian Eclipse v13.7) calculations of 730 fields in solid-water were compared to physical plan-specific quality assurance (PSQA) measurements acquired using a PTW Octavius 1500XDR array and PTW 31021 Semiflex 3D ion chamber.
Results:
TPS and MC showed good agreement with array measurements, evaluated using γ analyses at 3%, 3 mm with a 10% lower dose threshold:>94% of fields calculated by the TPS and >99% of fields calculated by MC had γ ≤ 1 for>95% of measurement points within the plane. TPS and MC also showed good agreement with chamber measurements of absolute dose, with systematic differences of <1.5% for all range-shifter options.
Conclusions:
Reliable independent verification of the TPS dose calculation is a valuable complement to physical PSQA and may facilitate reduction of the physical PSQA workload alongside a thorough delivery system quality assurance programme.
Advances In Knowledge:
A Gate/Geant4-based MC system is thoroughly validated against an extensive physical PSQA dataset for 730 clinical fields, showing that clinical implementation of MC for PSQA is feasible.
