Modeling physician's preference in treatment plan approval of stereotactic body radiation therapy of prostate cancer.
Modeling physician's preference in treatment plan approval of stereotactic body radiation therapy of prostate cancer.
复制标题
建模医师对治疗计划的偏爱批准了前列腺癌的立体定向身体放射治疗。
DOI：
10.1088/1361-6560/ac6d9e
复制
发表时间：
2022-05-26
影响因子：
3.5
通讯作者：
Jia, Xun
中科院分区：
文献类型：
--
作者：
Gao, Yin;Shen, Chenyang;Gonzalez, Yesenia;Jia, Xun
放射治疗的治疗计划是一项耗时的任务。期望开发自动规划方法以生成医生喜欢的规划。本研究的目的是开发一个基于深度学习的虚拟医生网络（VPN），模拟医生对前列腺癌立体定向体部放射治疗（SBRT）计划批准的偏好。VPN采用一个计划靶体积（PTV）和八个危险器官（OAR）结构图像，以及寻求批准的计划的剂量分布作为输入。它输出批准计划的概率和指示输入剂量的改进的剂量分布。由于我们的数据库中缺乏未经批准的计划，VPN使用对抗框架进行训练。本研究选择了接受45戈伊5次SBRT的68例前列腺癌患者，其中60例患者用于训练和交叉验证，8例患者用于独立测试。经过培训的VPN能够区分已批准和未批准的计划，测试数据的曲线下面积为0.97。对于未经批准的计划，应用VPN建议的剂量改进后，改进后的剂量与地面真实值的相对差异为：PTV D98%为2.03 ± 2.17%，PTV V95%为0.49 ± 0.29%，阴茎球Dmean为3.08 ± 2.24%，直肠V50%为3.73 ± 2.20%，膀胱V50%为2.06 ± 1.73%。VPN的开发是为了准确地模拟医生对计划批准的偏好，并提供有关如何改善剂量分布的建议。
Treatment planning of radiation therapy is a time-consuming task. It is desirable to develop automatic planning approaches to generate plans favorable by physicians. The purpose of this study is to develop a deep learning based virtual physician network (VPN) that models physician’s preference on plan approval for prostate cancer Stereotactic Body Radiation Therapy (SBRT). VPN takes one Planning Target Volume (PTV) and eight Organs at Risk (OAR) structure images, as well as a dose distribution of a plan seeking for approval as input. It outputs a probability of approving the plan, and a dose distribution indicating improvements to the input dose. Due to the lack of unapproved plans in our database, VPN is trained using an adversarial framework. 68 prostate cancer patients who received 45 Gy in 5-fraction SBRT were selected in this study, with 60 patients for training and cross validation, and 8 patients for independent testing. The trained VPN was able to differentiate approved and unapproved plans with Area under the curve 0.97 for testing data. For unapproved plans, after applying VPN’s suggested dose improvement, the improved dose agreed with ground truth with relative difference 2.03 ± 2.17% for PTV D98%, 0.49 ± 0.29% for PTV V95%, 3.08 ± 2.24% for penile bulb Dmean, 3.73 ± 2.20% for rectum V50%, and 2.06 ± 1.73% for bladder V50%. VPN was developed to accurately model a physician’s preference on plan approval and to provide suggestions on how to improve the dose distribution.
登录
查看更多内容
影响因子：
10.6
作者：
Mahmood F;Borders D;Chen RJ;Mckay GN;Salimian KJ;Baras A;Durr NJ
通讯作者：
Durr NJ
影响因子：
3.8
作者：
Shen C;Nguyen D;Chen L;Gonzalez Y;McBeth R;Qin N;Jiang SB;Jia X
通讯作者：
Jia X
DOI：
10.1186/s13014-015-0569-3
发表时间：
2015-12-29
期刊：
Radiation oncology (London, England)
影响因子：
--
作者：
Alfonso JC;Herrero MA;Núñez L
通讯作者：
Núñez L
DOI：
10.1016/j.ijrobp.2009.09.040
发表时间：
2010-03-01
影响因子：
7
作者：
Bentzen, Soren M.;Constine, Louis S.;Deasy, Joseph O.;Eisbruch, Avi;Jackson, Andrew;Marks, Lawrence B.;Ten Haken, Randall K.;Yorke, Ellen D.
通讯作者：
Yorke, Ellen D.
影响因子：
4.6
作者：
Dan Nguyen;Long, Troy;Jiang, Steve
通讯作者：
Jiang, Steve