A fast analytical dose calculation approach for MRI-guided proton therapy

Area of Science:

• Medical Physics
• Radiotherapy Technology
• Computational Imaging

Background:

• Magnetic resonance (MR) imaging offers innovative online image guidance for radiotherapy.
• Proton therapy is increasingly considered for MR-guided treatments, requiring fast dose calculations.
• Current Monte Carlo (MC) simulations for proton dose are accurate but time-consuming for online adaptation.

Purpose of the Study:

• To develop an efficient dose calculation method for MR-guided proton therapy.
• To create a fast GPU-based analytical algorithm incorporating magnetic field beam deflections.
• To enable rapid dose calculations essential for online plan adaptations.

Main Methods:

• A GPU-based analytical algorithm was developed, modifying ray casting dose calculation.
• Look-up tables (LUTs) of beam trajectories in orthogonal magnetic fields were generated using TOPAS-MC simulations.
• The algorithm was validated against MC simulations in various materials and patient cases.

Main Results:

• Excellent agreement between analytical and MC dose distributions was achieved (sub-millimetre range deviations, <2 mm lateral shifts).
• High gamma pass rates (2%/2 mm) were observed, comparable to 0 T scenarios, even in high-density materials.
• Comparable treatment plan quality was maintained regardless of magnetic field strength.

Conclusions:

• A novel, fast, and accurate method for proton dose calculation in magnetic fields has been developed.
• The algorithm is suitable for online plan adaptation in MR-guided proton therapy.
• This advancement facilitates improved treatment planning and delivery in advanced radiotherapy.