Author: Jacqueline M. Andreozzi, Savannah Decker, Diego Hernandez, Iman R. Washington 👨🔬
Affiliation: UC San Diego, Department of Radiation Oncology, Moffitt Cancer Center, University of South Florida, Moffitt Cancer Center 🌍
Purpose: Recent advances in external beam radiation therapy have begun to explore the utility of Cherenkov radiation imaging as a beam delivery visualization tool. When sheet bolus is used during treatment, its optical properties, along with the patient’s skin and tissue properties, will change the emission detected. To our knowledge, this is the first study to characterize the impact of optical signal generated in bolus coupled with the scattering and absorption properties of patient skin.
Methods: Optical phantom experiments were conducted by irradiating a silicone-based phantom embedded with flesh-colored pigments and translucent bolus (Elasto-Gel, thicknesses 3mm, 5mm, 10mm) with 6MV and 15MV photon beams. Opposing gantry angles were used to create entrance and exit beam geometries. This study also considered the impact of skin tone by using a thin, high-melanin-concentration interface layer between the bolus and optical phantom. We recorded a total of 24 measurements.
Results: The optical signal measured by the detector increases with increasing bolus thickness. Exit beam geometry yields greater average signal compared to the entrance beams, increasing with beam energy. Notably, the discrepancy between entrance and exit beam behavior is exacerbated when melanin is introduced. When 1.0 cm bolus is paired with melanin, the exit beam produces 10X the optical signal compared to when the same beam irradiates the phantom/melanin pair without bolus. This effect is attributed largely to optical photons generated in the bolus that are not absorbed by the melanin layer in this geometry.
Conclusion: The melanin content of skin affects the intensity observed by Cherenkov-based beam visualization tools, even when clinical bolus is used. Bolus was observed as generating its own optical photons in addition to the phantom-derived signal. Patient specific optical properties must be carefully considered, along with entrance versus exit beam, beam energy, and bolus thickness, when interpreting Cherenkov-based images.