[1]吴晓娟,张艺宝,任鸿儒,等.基于卷积神经网络的宫颈癌调强放疗剂量预测方法[J].中国医学物理学杂志,2025,42(4):421-428.[doi:10.3969/j.issn.1005-202X.2025.04.001]
　WU Xiaojuan,ZHANG Yibao,et al.Convolutional neural network based dose prediction method for intensity-modulatedradiotherapy of cervical cancer[J].Chinese Journal of Medical Physics,2025,42(4):421-428.[doi:10.3969/j.issn.1005-202X.2025.04.001]
点击复制

基于卷积神经网络的宫颈癌调强放疗剂量预测方法()
分享到：

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
42
期数:
2025年第4期
页码:
421-428
栏目:
医学放射物理
出版日期:
2025-04-20

文章信息/Info

Title:
Convolutional neural network based dose prediction method for intensity-modulatedradiotherapy of cervical cancer
文章编号:
1005-202X（2025）04-0421-08
作者:
吴晓娟 1; 2; 张艺宝 3; 任鸿儒 1; 孟令军 2
1.广东工业大学自动化学院，广东 广州 510000；2.山东大学齐鲁医院德州医院放射治疗科，山东 德州 253000；3.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
Author(s):
WU Xiaojuan1;  2;  ZHANG Yibao3;  REN Hongru1;  MENG Lingjun2
1. School of Automation, Guangdong University of Technology, Guangzhou 510000, China; 2. Department of Radiation Oncology,Dezhou Hospital, Qilu Hospital of Shandong University, Dezhou 253000, China; 3. Department of Radiation Oncology, PekingUniversity Cancer Hospital & Institute/Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education), Beijing100142, China
关键词:
宫颈癌; 剂量预测; 卷积神经网络; 调强放疗
Keywords:
cervical cancer;  dose prediction;  convolutional neural network;  intensity-modulated radiotherapy
分类号:
R318；R815.2；TP391.4
DOI:
10.3969/j.issn.1005-202X.2025.04.001
文献标志码:
A
摘要:
目的：建立一个卷积神经网络模型用于宫颈癌调强放疗的剂量分布，并评估它在实现自动计划方面的潜在应用。方法：采用100例女性盆腔调强放疗计划，80例为训练集，10例为验证集，10例为测试集，在三维深度残差网络模型基础上搭建剂量预测模型，预测三维剂量分布。通过CT 影像以及RT Structure文件提取出危及器官以及计划靶体积的掩模。将不同结构按照密度赋值图进行密度赋值，赋值后的CT掩图作为训练模型的输入图像。利用获取的最优模型实现三维剂量预测分布，将预测结果与手工计划设计的剂量分布进行剂量学比较。结果：在10例测试集上的实验结果显示，临床剂量学参数差异较小，在临床可接受范围内。10例测试集病例平均绝对误差 MAE为（0.58±0.16）Gy，平均 DSC系数为0.90±0.03，HD95为（10.61±7.17）mm。预测模型直肠的V45和小肠的D2 cc与手动计划相比，略有降低，左右侧股骨头的V20较手工计划有所降低；预测模型PTV的D95与人工计划相比有所降低，PTV的D90、HI和CI指数差异较小，两者的三维剂量分布差异较小，模型预测的剂量分布可满足临床要求。结论：基于卷积神经网络的剂量预测模型可以准确预测宫颈癌调强放疗的剂量分布，有望用于自动计划设计和质量评估等。
Abstract:
Objective To develop a convolutional neural network based model for predicting the dose distribution of intensitymodulated radiotherapy (IMRT) in cervical cancer, and to evaluate its potential applications in automated treatment planning.Methods The pelvic IMRT plans for 100 female patients were collected, with 80 cases in the training set, 10 in the validationset, and 10 in the test set. A dose prediction model was built based on the three-dimensional (3D) residual network forforecasting 3D dose distribution. Masks for organs-at-risk and planning target areas were extracted from CT images and RTStructure files. Density values were assigned to different structures according to a density map, and the resulting CT mapswere used as input images for model training. The optimal model was used to predict the 3D dose distribution, and thepredicted results were compared with the dose distribution from manual treatment planning in terms of dosimetricparameters. Results The experimental results on the 10-case test set demonstrated that dosimetric parameter differences wereinsignificant and within clinically acceptable ranges. The mean absolute error, average Dice similarity coefficient, and 95%Hausdorff distance for 10 cases in test set were (0.58±0.16) Gy, 0.90±0.03, and (10.61±7.17) mm, respectively. Comparedwith manual planning, prediction model showed slightly decreased rectal V45, small bowel D2 cc, and the V20of bilateralfemoral heads was reduced. The predicted D95 of planning target area was lower than manual planning, but the differences inD90, homogeneity index, and conformity index were trivial. There were minor differences in 3D dose distributions betweenthe two, and the dose distribution generated by prediction model met clinical requirements. Conclusion The convolutional neural network based dose prediction model can accurately forecast the dose distribution for cervical cancer IMRT, exhibitingthe potential to be used in automated treatment planning and quality evaluation.

相似文献/References:

[1]戴相昆,杨丰蔚,杜 镭,等.等效均匀剂量优化法在宫颈癌调强计划中的应用[J].中国医学物理学杂志,2015,32(02):290.[doi:10.3969/j.issn.1005-202X.2015.02.033]
[2]刘 浩,王 新,李公平,等.宫颈癌放疗高危临床靶区变化及其对累计吸收剂量的影响[J].中国医学物理学杂志,2015,32(03):379.[doi:10.3969/j.issn.1005-202X.2015.03.017]
[3]孙晓欢,周咏春,谭丽娜,等.容积旋转调强与固定野调强在宫颈癌术后放疗中的剂量学比较[J].中国医学物理学杂志,2014,31(01):4604.[doi:10.3969/j.issn.1005-202X.2014.01.003]
[4]刘 浩,王 新,李公平,等.用二维电离室矩阵验证宫颈癌腔内放疗剂量分布[J].中国医学物理学杂志,2014,31(04):4988.[doi:10.3969/j.issn.1005-202X.2014.04.004]
[5]余 辉,张书旭.宫颈癌后装治疗剂量的调强优化方法[J].中国医学物理学杂志,2014,31(05):5114.[doi:10.3969/j.issn.1005-202X.2014.05.004]
[6]查元梓,张松方,屠永清,等.子野权重优化对CMS XiO宫颈癌调强计划的影响[J].中国医学物理学杂志,2013,30(01):3866.[doi:10.3969/j.issn.1005-202X.2013.01.007]
[7]张富利,刘清智,张伟超,等.宫颈癌术后盆腔螺旋断层调强放疗两种计划设计方法的剂量学分析[J].中国医学物理学杂志,2013,30(04):4229.[doi:10.3969/j.issn.1005-202X.2013.04.003]
[8]王佳浩,赖建军,唐荣军,等.结合EPID与EUD方法确定宫颈癌病例靶区最佳生物外放边界的研究[J].中国医学物理学杂志,2013,30(05):4378.[doi:10.3969/j.issn.1005-202X.2013.05.009]
[9]张富利,刘清智,王雅棣,等.基于配置不同类型MLC叶片加速器的宫颈癌术后调强放疗计划评估[J].中国医学物理学杂志,2013,30(05):4383.[doi:10.3969/j.issn.1005-202X.2013.05.010]
[10]刘致滨,王占宇,龙雨松.两种优化技术在宫颈癌术后IMRT计划中的比较[J].中国医学物理学杂志,2015,32(04):534.[doi:10.3969/j.issn.1005-202X.2015.04.018]
　[J].Chinese Journal of Medical Physics,2015,32(4):534.[doi:10.3969/j.issn.1005-202X.2015.04.018]

备注/Memo

备注/Memo:
【收稿日期】2024-12-19【基金项目】国家自然科学基金（U21A20522，12275012，12475309，12411530076）；北京市自然科学基金（Z210008）【作者简介】吴晓娟，硕士，研究方向：智能医学工程，E-mail: wjuan16@163.com【通信作者】孟令军，副主任技师，研究方向：放射物理，E-mail: 13605349565@163.com
更新日期/Last Update: 2025-04-30