Suppr超能文献

利用 Timepix 探测器和人工智能对单质子线性能进行特征描述,以用于先进的质子治疗计划。

Single proton LET characterization with the Timepix detector and artificial intelligence for advanced proton therapy treatment planning.

机构信息

Institute of Nuclear Physics Polish Academy of Sciences, PL-31342 Krakow, Poland.

Baylor University, Waco, TX 76706, Texas, United States of America.

出版信息

Phys Med Biol. 2023 May 8;68(10). doi: 10.1088/1361-6560/acc9f8.

Abstract

Protons have advantageous dose distributions and are increasingly used in cancer therapy. At the depth of the Bragg peak range, protons produce a mixed radiation field consisting of low- and high-linear energy transfer (LET) components, the latter of which is characterized by an increased ionization density on the microscopic scale associated with increased biological effectiveness. Prediction of the yield and LET of primary and secondary charged particles at a certain depth in the patient is performed by Monte Carlo simulations but is difficult to verify experimentally.Here, the results of measurements performed with Timepix detector in the mixed radiation field produced by a therapeutic proton beam in water are presented and compared to Monte Carlo simulations. The unique capability of the detector to perform high-resolution single particle tracking and identification enhanced by artificial intelligence allowed to resolve the particle type and measure the deposited energy of each particle comprising the mixed radiation field. Based on the collected data, biologically important physics parameters, the LET of single protons and dose-averaged LET, were computed.An accuracy over 95% was achieved for proton recognition with a developed neural network model. For recognized protons, the measured LET spectra generally agree with the results of Monte Carlo simulations. The mean difference between dose-averaged LET values obtained from measurements and simulations is 17%. We observed a broad spectrum of LET values ranging from a fraction of keVmto about 10 keVmfor most of the measurements performed in the mixed radiation fields.It has been demonstrated that the introduced measurement method provides experimental data for validation of LETor LET spectra in any treatment planning system. The simplicity and accessibility of the presented methodology make it easy to be translated into a clinical routine in any proton therapy facility.

摘要

质子具有有利的剂量分布,并且越来越多地用于癌症治疗。在布拉格峰深度范围内,质子产生由低和高线性能量转移(LET)成分组成的混合辐射场,后者的特征是微观尺度上的电离密度增加,与生物效应增加相关。在患者体内的特定深度处预测初级和次级带电粒子的产额和 LET 通过蒙特卡罗模拟进行,但难以通过实验验证。这里,呈现了在水中治疗质子束产生的混合辐射场中使用 Timepix 探测器进行的测量结果,并与蒙特卡罗模拟进行了比较。探测器通过人工智能进行高分辨率单粒子跟踪和识别的独特能力,允许解析粒子类型并测量构成混合辐射场的每个粒子的沉积能量。基于收集的数据,计算了生物学上重要的物理参数,即单质子的 LET 和剂量平均 LET。使用开发的神经网络模型,质子识别的准确率超过 95%。对于识别出的质子,测量的 LET 谱通常与蒙特卡罗模拟的结果一致。从测量和模拟获得的剂量平均 LET 值之间的平均差异为 17%。我们观察到,在混合辐射场中进行的大多数测量中,LET 值的范围很广,从几分之一 keVm 到大约 10 keVm。已经证明,所提出的测量方法提供了任何治疗计划系统中 LET 或 LET 谱的验证的实验数据。所提出的方法的简单性和可访问性使其易于在任何质子治疗设施中转化为临床常规。

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验