Presenter(s)
W. Zhang1, J. Park2, I. Oraiqat3, G. Gonzalez4, Y. Liu1, Y. Huang1, S. Dykstra1, L. Wei5, D. W. Litzenberg5, K. C. Cuneo5, W. M. Mendenhall2, S. Jean-Baptiste2, C. M. Bryant2, P. Johnson2, I. El Naqa4, and X. Wang1; 1University of Michigan, Ann Arbor, MI, 2University of Florida Health Proton Therapy Institute, Jacksonville, FL, 3H. Lee Moffitt Cancer Center and Research Institute, Department of Radiation Oncology, Tampa, FL, 4H. Lee Moffitt Cancer Center and Research Institute, Department of Machine Learning, Tampa, FL, 5Department of Radiation Oncology, University of Michigan, Ann Arbor, MI
Purpose/Objective(s): Proton beam therapy (PBT) offers significant advantages in cancer treatment by precisely targeting tumors while sparing surrounding healthy tissues. However, the accuracy of dose delivery is often limited by range uncertainties and anatomical changes during treatment. Despite technical advances in PBT, real-time and precise localization of the Bragg peak inside the patient remains a major challenge, with no reliable solutions currently available in clinical practice. This study aims to develop and validate a method based on ionizing radiation-induced acoustic effects to localize the proton beam delivery spot in real time with high sensitivity, high resolution, and high accuracy.
Materials/Methods: An ionizing radiation acoustic beam localization (iRABL) approach was proposed to achieve surgical-level precision in proton beam localization. A specialized iRABL system was developed with high-speed, high-resolution, and high-sensitivity capabilities tailored for proton beam therapy applications. The clinical feasibility of the system was validated in vivo through the first human study involving prostate cancer patients receiving PBT. Performance metrics—including displacement resolution, imaging and processing speed, and accuracy for tracking pencil beam scanning trajectories and temporal dose accumulation—were evaluated using soft tissue phantoms and clinical treatment plan of patients imaged with radiation acoustic system.
Results: The first-in-human study demonstrated that iRABL successfully provides real-time imaging and accurate tracking of proton beam delivery during prostate cancer PBT. The system achieved a displacement resolution of 0.1 mm laterally and 0.2 mm axially, exceeding both the diffraction limit of proton-induced acoustic waves and the physical constraints of the proton beam. Leveraging GPU acceleration, the system demonstrated single-pulse temporal resolution and tracked at a frame rate greater than 1 kHz, confirming its exceptional sensitivity and speed. The imaging accuracy of iRABL for M-shaped treatment plans achieve a gamma index of 92% with 2mm/3% criteria, validating its precision in tracking pencil beam delivery in clinical settings.
Conclusion: The iRABL method enables real-time tracking of temporal dose accumulation and pencil beam scanning trajectories with submillimeter spatial resolution, single-pulse temporal resolution, and high sensitivity during PBT. These findings highlight iRABL’s potential as a practical tool for real-time treatment guidance, advancing PBT toward precision comparable to surgical standards.