Hybrid Monte Carlo model for efficient tissue Cherenkov emission estimation to assess changes from beam size, energy...

Cherenkov imaging in radiation therapy provides key knowledge of the delivery of treatment plans, but light-tissue interactions alter the emitted spectral signal and cause the modeling of emission relative to dose in highly modulated treatment plans to be complex.

A 2-stage Monte Carlo approach to modeling Cherenkov emission was developed that leverages a traditional treatment planning system with an optical Monte Carlo simulation to provide a widely useable and efficient tool for modeling every beam control point for delivery interpretation of highly-modulated treatment plans. The emitted optical spectra were estimated for 6, 10, 15MV photon beams, 6 MeV electron beams, beam incidence in tissue, and square field sizes from 1 cm to 20 cm. The model was validated through comparison of measured Cherenkov emission from a blood and intralipid optical phantom.

The resulting hybrid model provides an efficient method of estimating Cherenkov emission for linac beams, showing a clear trend of decreasing emission intensity with increasing beam energy and strong emission intensity variation with beam type. The largest change in observed intensity was from altering field size, with a 76 % intensity decrease when going from 20 cm down to 1 cm square. The model showed agreement with experimental detected Cherenkov with an average percent difference of 6.2 % with the largest difference at the very smallest beam sizes.

The model potentially allows for modeling entire modulated treatment plans with high computational efficiency and is a key step to translate delivered dose and observed Cherenkov in highly modulated situations.