Quantitative Cherenkov-excited fluorescence emission dosimetry for HDR Co-60 brachytherapy: Calibration and planar...

Accurate dose verification in High-Dose-Rate (HDR) brachytherapy presents a significant challenge due to steep dose gradients and the absence of beam collimation. Conventional quality assurance (QA) methods typically rely on point detectors or film-based techniques, which lack the capability to provide real-time, full-field visualization of dose distributions in a water-equivalent medium. While Cherenkov emission (CE) imaging has been investigated for Iridium-192 (¹⁹²Ir) systems, the low mean photon energy of ¹⁹²Ir (∼380 keV) constrains the optical yield and limits the signal-to-noise ratio. Because Cobalt-60 (⁶⁰Co) emits higher primary photon energies, it generates distinct Cherenkov intensities and spatial characteristics, requiring dedicated evaluation.

This study aims to characterize the relationship between Cherenkov emission intensity and absorbed dose generated by an HDR ⁶⁰Co source. Furthermore, it seeks to generate high-resolution two-dimensional dose maps using Cherenkov-excited fluorescence imaging and validate their dosimetric accuracy against Treatment Planning System (TPS) calculations to evaluate the system for clinical QA implementation.

A custom 3D-printed polylactic acid (PLA) phantom was fabricated with an internal cavity filled with an aqueous quinine hemisulfate solution (1 g/L). This solution was used to convert ultraviolet Cherenkov photons into isotropic visible fluorescence, thereby enhancing optical signal detection. Optical emissions were captured using a standard monochrome CMOS camera positioned at a 24 cm working distance. To eliminate radiation-induced artifacts on the sensor without relying on complex time-gated hardware, an image processing pipeline comprising dark frame subtraction, median filtering, and morphological operations was applied. A calibration function was established by delivering doses between 10 and 1500 cGy using a ⁶⁰Co HDR afterloader. Dosimetric accuracy was then evaluated across ten independent verification plans using two-dimensional gamma index analysis with 3% dose difference and 3 mm distance-to-agreement (3%/3 mm) criteria.

The dosimetric calibration demonstrated a highly predictable and linear dose-response relationship, yielding a coefficient of determination (R²) of 0.9999 and a root mean square error (RMSE) of 3.46 cGy. For the clinical verification plans, the CE-derived 2D dose maps exhibited strong spatial agreement with the TPS calculations. Gamma passing rates across the ten plans ranged from 91.6% to 97.3%, resulting in a mean passing rate of 94.2% ± 1.9% and an average mean gamma value of 0.57 ± 0.05. One-dimensional line profiles further confirmed the high spatial fidelity of the system in accurately tracking steep dose fall-offs.

The integration of Cherenkov-excited fluorescence imaging with a standard CMOS sensor provides a highly linear, practical, and cost-effective planar dose verification tool for continuous-emission ⁶⁰Co HDR brachytherapy. The proposed methodology mitigates the signal limitations inherent to lower-energy isotopes and achieves strong spatial agreement with TPS calculations, supporting its reliability as an independent QA modality in a controlled phantom environment.