Dose kernel decomposition for spot-based radiotherapy treatment planning.

Area of Science:

• Medical Physics
• Radiotherapy Physics
• Computational Dosimetry

Background:

• Pre-calculation of dose deposition kernels for spot-based radiotherapy, like Gamma Knife (GK) and Gamma Pod (GP), is computationally intensive and requires substantial data storage.
• Existing methods for dose calculation in GK and GP can be time-consuming, hindering efficient treatment planning.

Purpose of the Study:

• To introduce and validate a novel kernel decomposition (KD) model for accurate and rapid dose calculation in spot-based radiotherapy.
• To reduce data storage requirements and enable real-time dose calculations for GK and GP treatment planning.

Main Methods:

• Modeled dose deposition kernels as a product of a shift-invariant reference kernel and a spatially variant scale factor.
• Calculated reference kernels using Monte Carlo (MC) methods and scale factors derived from tissue maximum ratios (TMR).
• Validated the KD model against MC simulations and treatment planning system (TPS) calculations for phantom and clinical GK and GP plans.

Main Results:

• KD model achieved >99% Gamma index passing rates (3%/1mm) compared to MC for phantom and clinical GK plans.
• KD model demonstrated >95% Gamma index passing rates compared to MC for clinical GP plans.
• Scale factor and dose kernel calculations were completed in sub-seconds, significantly reducing computation time and data storage needs.

Conclusions:

• The KD model provides accurate, real-time dose and derivative calculations for GK and GP radiotherapy.
• This approach eliminates the need for pre-calculation and large data storage of dose kernels.
• The KD model facilitates efficient treatment planning for spot-based radiotherapy by enabling rapid optimization of spot selection.