Polarized Cherenkov light imaging dosimetry: the impact of source to detector distance

<i>Objective.</i>With every treatment vault being different, the impact of geometrical parameters such as the signal source-to-camera distance on dose proportionality must be evaluated. The aim of this study is to characterize the Cherenkov signal and polarization state as a function of the source-to-camera distance. Our hypothesis is that with increasing distance, the need for angular correction distributions decreases, resulting in acquisition of a polarized Cherenkov signal directly proportional to dose.<i>Approach.</i>A water tank and a polyvinyltoluene-based plastic scintillator volume were irradiated by a 6 MV beam to respectively produce Cherenkov emissions as well as a control signal. Monte Carlo reference simulations were performed using TOPAS. Acquisitions of the Cherenkov signal were achieved using a cooled CCD camera and a time-gated intensified CMOS camera. By fitting a modified Malus Law to the Cherenkov acquisitions, the total Cherenkov signal intensity and its purely polarized component was extracted. Signal source-to-camera distance of 0.5, 1, 2, 3 and 4 m were tested to evaluate this distance's impact on the signal distributions. Projected percent depth dose (PPDD) and projected transverse profiles calculated from the different signal sources were then compared.<i>Main results.</i>All PPDDs at camera distances of 3 and 4 m agree with Monte Carlo (⩽5%) over depths ranging from 1.5 to 16 cm. Cherenkov PPDDs at camera distances of 0.5 and 1 m show significant discrepancies (⩾5%) compared to MC because no angular corrections are applied. Over the plateau region of projected transverse profiles, general agreement with MC is achieved. Thirteen of the 17 luminescence-based beam widths show⩽5% differences with MC.<i>Significance.</i>This study confirms the above-mentioned hypothesis up until the image quality diminishes. For this work's setup, the optimal camera distance for dosimetry using Cherenkov polarized imaging was found to be between 3 and 4 m.