A. L. Matous1, S. M. Decker2, R. Zhang2, D. J. Gladstone3, E. K. Grove4, B. B. Williams5, M. Jermyn6, S. M. McVorran7, and L. A. Jarvis8; 1Dartmouth-Hitchcock Medical Center, Lebanon, NH, United States, 2Thayer School of Engineering, Dartmouth College, Hanover, NH, 3Geisel School of Medicine at Dartmouth & Norris Cotton Cancer Center, Dartmouth Hitchcock Medical Center, Lebanon, NH, 4Dartmouth Health, Lebanon, NH, 5Department of Medicine, Radiation Oncology, Dartmouth-Hitchcock Medical Center / Norris Cotton Cancer Center, Lebanon, NH, 6DoseOptics, LLC, Lebanon, NH, 7H. Lee Moffitt Cancer Center and Research Institute, Department of Radiation Oncology, Tampa, FL, 8Department of Radiation Oncology and Applied Sciences, Dartmouth Cancer Center, Dartmouth Health, Lebanon, NH
Purpose/Objective(s): Cardiac implanted electronic devices (CIED) require dose monitoring during each fraction of radiotherapy. Current methodologies can be time consuming and may have delayed read-out times, making them inconvenient for routine clinical use. We hypothesize that Cherenkov imaging combined with scintillation dosimetry can be an efficient, accurate, real-time, on-patient verification system for monitoring radiotherapy delivery to patients with CIEDs. Materials/
Methods: A Cherenkov imaging system consisting of two time-gated, iCMOS cameras was used to collect video images of both an anthropomorphic chest phantom with a CIED affixed on the surface, as well as patients undergoing radiation treatment near chest wall cardiac devices. Positional accuracy was assessed using a predicted surface dose overlay. To record absolute dosimetry, plastic scintillator discs were placed at points of interest on both the phantom and patients. Discs were placed alongside optically stimulated luminescence dosimeters (OSLDs) for dose measurement comparisons. Results: In phantom studies, Cherenkov images visibly indicated when dose was delivered to the cardiac device as compared to non-overlapping dose deliveries. The accuracy of spatial delivery was further assessed by overlaying the predicted surface dose outline derived from the treatment planning system (TPS) with the Cherenkov images. The comparison revealed congruence at the planned position and discrepancies (non-congruence) when the phantom was shifted from the initial position. Finally, the absolute doses derived from the scintillator discs in the phantom study aligned well with both the OSLD measurements and TPS predictions for three different positional cases, measuring within 10% for in-field positions and within 5% for out of field positions. To demonstrate in vivo applicability of this method, patients with cardiac devices were imaged over 18 daily fractions. Surface plan comparisons confirmed the positional accuracy of dose delivery for all fractions. Daily dose delivered to the cardiac devices, as measured by scintillator discs, deviated by no more than 1.5 cGy from the OSLD measurements. Conclusion: Cherenkov imaging combined with scintillation dosimetry presents an alternative solution to current methodologies for CIED monitoring. This approach has the benefit of instantly detecting deviations, enabling timely corrective measures and proper patient triage.