In-Beam PET Imaging with Depth-Of-Interaction Measurement for Accurate Proton Beam-Range Verification
Y Shao1*, X Sun1, K Lou1,2, X Zhu1, D Mirkovic1, F Poenisch1, D Grosshans1, (1) The University of Texas M.D. Anderson Cancer Ctr., Houston, TX, (2) Rice University, Houston, TexasWE-G-500-3 Wednesday 4:30PM - 6:00PM Room: 500 Ballroom
Purpose: To study the feasibility of using an in-beam PET with depth-of-interaction (DOI) measurement to improve beam-range verification for proton therapy.
Methods: A compact prototype PET with dual rotating DOI-measurable detector panels was developed based on using solid-state photomultiplier (SSPM) arrays to read LYSO scintillators with parallel readout ASIC electronics. The system had a 44mm diameter trans-axial and 30mm-axial field of view (FOV). A 36mm diameter PMMA phantom was placed inside FOV. Both PET and phantom axes were aligned with collimated 200 MeV beams. The Bragg peak was located within the axial FOV by adjusting phantom position. A total of 9 beams irradiated the phantom with a 20-40 minutes beam separation; each beam delivered ~50 spills (0.5 sec spill and 1.5 sec inter-spill time, 800 MU); data from each beam were acquired with detectors at a certain angle; 9 datasets for 9 beams with detectors at 9 different angles over 180o were acquired; each dataset collected both in-beam and 5 min after-beam data. Beam-range was measured from the PET image reconstructed from all 9 datasets, and compared to the results from simulated images. Additionally, a Na-22 disk-source was also acquired after each beam to measure the impact of neutrons on system performance.
Results: PET performed well except energy photo-peak positions were reduced slightly after each beam, presumably secondary to neutron exposure of the SSPM. This minor effect was corrected with a shifting 350-650 keV energy window for each dataset. The difference between measured and simulated beam-ranges was within 1.0 mm, and this can be achieved with in-beam data alone. DOI-measurement provided uniform resolutions and high sensitivity to improve the accuracy of range verification.
Conclusion: A SSPM-based DOI-measureable PET capable of in-beam imaging may offer significant improvements in the accuracy of proton beam-range verification.
Funding Support, Disclosures, and Conflict of Interest: This project is supported by award RP120326 from the Cancer Prevention & Research Institute of Texas.