The School of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan, People's Republic of China.
Cancer Center, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, People's Republic of China.
Phys Med Biol. 2024 Jun 7;69(12). doi: 10.1088/1361-6560/ad4f43.
. In-beam positron emission tomography (PET) is a promising technology for real-time monitoring of proton therapy. Random coincidences between prompt radiation events and positron annihilation photon pairs can deteriorate imaging quality during beam-on operation. This study aimed to improve the PET image quality by filtering out the prompt radiation events.. We investigated a prompt radiation event filtering method based on the accelerator radio frequency phase and assessed its performance using various prompt gamma energy thresholds. An in-beam PET prototype was used to acquire the data when the 70 MeV proton beam irradiated a water phantom and a mouse. The signal-to-background ratio (SBR) indicator was utilized to evaluate the quality of the PET reconstruction image.. The selection of the prompt gamma energy threshold will affect the quality of the reconstructed image. Using the optimal energy threshold of 580 keV can obtain a SBR of 1.6 times for the water phantom radiation experiment and 2.0 times for the mouse radiation experiment compared to those without background removal, respectively.. Our results show that using this optimal threshold can reduce the prompt radiation events, enhancing the SBR of the reconstructed image. This advancement contributes to more accurate real-time range verification in subsequent steps.
. 在线正电子发射断层扫描(PET)是质子治疗实时监测的一项很有前途的技术。在束流照射过程中,快辐射事件与正电子湮没光子对之间的随机符合会降低成像质量。本研究旨在通过过滤掉快辐射事件来提高 PET 图像质量。. 我们研究了一种基于加速器射频相位的快辐射事件过滤方法,并使用各种快伽马能量阈值评估了其性能。使用在线 PET 原型机在水模和小鼠上辐照 70 MeV 质子束时获取数据。使用信号与背景比(SBR)指标评估 PET 重建图像的质量。. 快伽马能量阈值的选择会影响重建图像的质量。使用最佳能量阈值 580 keV 可以分别使水模辐射实验和小鼠辐射实验的 SBR 提高到未去除背景的 1.6 倍和 2.0 倍。. 我们的结果表明,使用这个最佳阈值可以减少快辐射事件,提高重建图像的 SBR。这一进展有助于后续步骤中更准确的实时范围验证。