中华放射医学与防护杂志

胡慧敏,董正坤,余疏桐,等.基于双能锥束CT和深度学习的放疗在体剂量监测基准合成方法[J].中华放射医学与防护杂志,2025,45(2):129-136.Hu Huimin,Dong Zhengkun,Yu Shutong,et al.A method to establish reference benchmarks for in vivo dose monitoring for radiotherapy based on dual-energy cone beam CT and deep learning[J].Chin J Radiol Med Prot,2025,45(2):129-136

基于双能锥束CT和深度学习的放疗在体剂量监测基准合成方法

A method to establish reference benchmarks for in vivo dose monitoring for radiotherapy based on dual-energy cone beam CT and deep learning

投稿时间：2024-08-24

DOI：10.3760/cma.j.cn112271-20240824-00319

中文关键词: 在线自适应放疗在体剂量监测双能锥束CT 对抗生成网络

英文关键词:Online adaptive radiotherapy In vivo dose monitoring Dual-energy cone-beam computed temography Generative adversarial network (GAN)

基金项目:国家自然科学基金(12475309,12275012,12411530076,82202941);北京市自然科学基金(Z210008);中央高校基本科研业务费/北京大学临床医学+X青年专项(PKU2024LCXQ033);教育部内地与港澳高等学校师生交流计划项目(万人计划7111400049);国家重点研发计划(2019YFF01014405);内蒙古自治区科技计划(2022YFSH0064)

作者	单位	E-mail
胡慧敏	北京大学医学部医学技术研究院, 北京 100191
董正坤	北京大学未来技术学院, 北京 100871
余疏桐	北京大学医学部医学技术研究院, 北京 100191
林晨	北京大学核物理与核技术国家重点实验室, 北京 100871
黎田	香港理工大学医疗科技及资讯学系, 香港 999077
张艺宝	北京大学医学部医学技术研究院, 北京 100191 北京大学肿瘤医院暨北京市肿瘤防治研究所放疗科, 恶性肿瘤发病机制及转化研究教育部重点实验室, 北京 100142	zhangyibao@pku.edu.cn

摘要点击次数: 1274

全文下载次数: 498

中文摘要:

目的利用改进的CycleGAN网络实现从双能kV锥形束CT(DECBCT)到兆伏级(MV)投影的转换, 为在线自适应放疗等先进技术的安全实施提供潜在可用的出射束在体剂量参考基准和实时监测方法。方法利用4D扩展心脏躯干(XCAT)模型生成模拟患者数据, 基于Varian机载锥束CT的几何参数生成投影, 并通过迭代双能分解算法从DECBCT图像中获取相对电子密度(RED)图像。分别以DECBCT投影和RED图像的投影作为输入, 训练SE-CycleGAN和CycleGAN网络用于生成MV投影图像。利用结构相似性指数(SSIM)、峰值信噪比(PSNR)、均方根误差(RMSE)等指标评估各方法的有效性。结果 SE-CycleGAN生成结果的全部评价指标显著优于CycleGAN(Z = -23.92、-26.17、-25.54、-26.80、-11.54、-11.21, P <0.05), 尤其在全局信息的学习方面。此外, 使用DECBCT投影图像作为输入的训练效果优于使用RED图像, 尽管两者均能生成较为理想的MV投影图像。在测试集的全部3 636组投影中:以DECBCT投影作为输入的SE-CycleGAN网络和CycleGAN网络的评估结果分别为:0.997 7±0.000 7、0.997 1±0.001 6 (SSIM);39.625 0±4.684 4、36.272 2±5.566 3 (PSNR);0.004 1±0.002 7、0.006 3±0.004 3 (RMSE)。以RED投影作为输入的SE-CycleGAN网络和CycleGAN网络的评估结果分别为:0.996 8±0.001 0、0.996 2±0.001 5(SSIM);38.548 7±3.637 4、36.007 3±4.437 8 (PSNR);0.004 3±0.002 2、0.006 1±0.003 7(RMSE)。结论本研究利用双能锥束CT和深度学习技术, 提出了建立在体剂量监测参考基准的新方法, 并基于虚拟仿真实验初步证明了该方法的准确性和有效性。

英文摘要:

Objective To achieve the conversion from dual-energy cone-beam CT (DECBCT) at the kilovolt (KV) level to projections at the megavolt (MV) level using an improved CycleGAN network, in order to provide a potential reference benchmark and real-time monitoring of in vivo doses delivered by exit beams for the safe implementation of advanced techniques such as online adaptive radiotherapy. Methods Simulated patient data were generated using a 4D extended cardiac torso (XCAT) model, and projections were generated based on the geometric parameters of Varian's onboard cone-beam CT. Furthermore, relative electron density (RED) images were derived from DECBCT images using an iterative dual-energy decomposition algorithm. The SE-CycleGAN and CycleGAN networks were trained to generate MV projection images using DECBCT projections and RED images, respectively. The performance of both methods was evaluated using metrics including structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), and root mean square error (RMSE). Results SE-CycleGAN significantly outperformed CycleGAN in all evaluation metrics (Z = -23.92, -26.17, -25.54, -26.80, -11.54, -11.21, P < 0.05), particularly in learning global information. Besides, although both methods generated satisfactory MV projections, training using DECBCT projections as input yielded better effects than training using RED images. For all the 3 636 sets of projections in the test set, the SE-CycleGAN and CycleGAN networks using DECBCT projections as input respectively yielded SSIMs of 0.9977±0.0007 and 0.9971±0.0016, PSNRs of 39.6250±4.6844 and 36.2722±5.5663, and RMSEs of 0.0041±0.0027 and 0.0063±0.0043, respectively. In contrast, the SE-CycleGAN and CycleGAN networks using RED projections as input respectively yielded SSIMs of 0.9968±0.0010 and 0.9962±0.0015, PSNRs of 38.5487±3.6374 and 36.0073±4.4378, and RMSEs of 0.0043±0.0022 and 0.0061±0.0037, respectively. Conclusions This study proposed a new method to establish reference benchmarks for in vivo dose monitoring based on DECBCT and deep learning technologies. This method is accurate and effective according to the preliminary validation using virtual simulation experiments.

HTML 查看全文查看/发表评论下载PDF阅读器

关闭