Material assignment for proton range prediction in Monte Carlo patient simulations using stopping-power datasets

Area of Science:

• Medical Physics
• Computational Biology
• Radiotherapy

Background:

• Proton therapy patient simulation relies on accurate material property assignment.
• Current methods using computed tomography (CT) are dependent on scanner-specific settings, posing a calibration challenge.
• A universal approach is needed for consistent and accurate Monte Carlo (MC) simulations.

Purpose of the Study:

• To introduce and validate a novel, universally applicable material assignment approach (MATA) for MC patient simulation in proton therapy.
• To demonstrate MATA's independence from CT acquisition and reconstruction parameters.
• To evaluate the impact of MATA on dose distribution and proton range prediction.

Main Methods:

• MATA assigns material properties to the stopping-power ratio (SPR) using 40 human tissue compositions and mass density.
• Clinically available CT-number-to-SPR conversion is utilized, avoiding further calibration.
• Validation involved homogeneous/heterogeneous SPR datasets and application to patient data.

Main Results:

• MATA achieved near-zero deviation in SPR for homogeneous datasets and <0.2% for heterogeneous datasets.
• Significant differences in dose distribution and SPR were observed when comparing MATA to other assignment methods.
• Patient-specific proton range shifts varied from 1.3 mm to 4.8 mm between different SPR prediction approaches.

Conclusions:

• MATA offers a universal solution for patient modeling in MC-based proton treatment planning, removing the need for institution-specific adaptations.
• The approach facilitates the integration of more accurate SPR prediction methods.
• MATA enhances the consistency and accuracy of proton therapy simulations across different institutions.