Feasibility of using megavoltage computed tomography to reduce proton range uncertainty: A simulation study

Area of Science:

• Medical Physics
• Radiotherapy Physics
• Computational Imaging

Background:

• Accurate proton stopping power ratio (SPR) estimation is crucial for precise proton therapy planning.
• Kilovoltage dual-energy computed tomography (kVCT) is currently used, but its accuracy can be affected by tissue chemical composition variations.
• Megavoltage computed tomography (MVCT) offers a potential alternative due to its different scattering properties.

Purpose of the Study:

• To assess the impact of chemical composition variations on the relationship between relative electron density (RED) and proton SPR.
• To establish a theoretical basis for utilizing MVCT, instead of kVCT, for accurate proton SPR estimation.
• To evaluate the feasibility of using RED as a direct predictor of proton SPR.

Main Methods:

• A simulation study was conducted using reference and simulated human tissues with varying chemical compositions.
• Proton SPR and kVCT numbers were theoretically calculated and estimated from RED using stoichiometric calibration curves.
• Estimation errors were quantified by comparing calculated and RED-derived values.

Main Results:

• Chemical composition variations of 5% (1σ) resulted in minimal estimation errors for kVCT number (0.34%-0.77%) and proton SPR (0.16%-0.30%) across different tissue types.
• Relative electron density (RED) alone proved sufficient for accurate proton SPR determination.
• A strong linear correlation was observed between MVCT number and RED, attributed to Compton scattering dominance.

Conclusions:

• Tissue chemical composition has a negligible effect on determining proton SPR using RED.
• RED is a reliable metric for accurately estimating proton SPR.
• MVCT's linear relationship with RED suggests its potential for improving proton range accuracy and reducing uncertainties in radiotherapy.