[1]熊互湟,张盛元,王诗露,等.基于Transformer与门控交互特征融合的大核分解卷积网络用于宫颈癌靶区自动勾画[J].中国医学物理学杂志,2026,43(9):1267-1274.[doi:DOI:10.3969/j.issn.1005-202X.2026.09.015]
　XIONG Huhuang,ZHANG Shengyuan,WANG Shilu,et al.Large-kernel separable convolutional neural network based on Transformer and gated interactive feature fusion for automatic segmentation of planning target volume in cervical cancer[J].Chinese Journal of Medical Physics,2026,43(9):1267-1274.[doi:DOI:10.3969/j.issn.1005-202X.2026.09.015]
点击复制

基于Transformer与门控交互特征融合的大核分解卷积网络用于宫颈癌靶区自动勾画()

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

卷:
43卷
期数:
2026年第9期
页码:
1267-1274
栏目:
医学人工智能
出版日期:
2026-09-30

文章信息/Info

Title:
Large-kernel separable convolutional neural network based on Transformer and gated interactive feature fusion for automatic segmentation of planning target volume in cervical cancer
文章编号:
1005-202X（2026）09-1267-08
作者:
熊互湟1; 张盛元2; 王诗露1; 勾成俊1; 吴章文1
1.四川大学原子核科学技术研究所/辐射物理与技术教育部重点实验室， 四川 成都 610064；2.陕西省肿瘤医院放疗科， 陕西 西安 710061
Author(s):
XIONG Huhuang1;  ZHANG Shengyuan2;  WANG Shilu1;  GOU Chengjun1;  WU Zhangwen1
1. Key Laboratory of Radiation Physics and Technology, Ministry of Education/Institute of Nuclear Science and Technology, Sichuan University, Chengdu 610064, China 2. Department of Radiation Oncology, Shaanxi Provincial Cancer Hospital, Xian 710061, China
关键词:
深度学习; 医学图像分割; 宫颈癌; 放射治疗
Keywords:
Keywords: deep learning medical image segmentation cervical cancer radiotherapy
分类号:
R318；R811.1
DOI:
DOI:10.3969/j.issn.1005-202X.2026.09.015
文献标志码:
A
摘要:
针对放疗计划中靶区勾画依赖人工操作、效率低且一致性差的问题，提出一种基于Transformer结构的高精度自动分割模型Trans-GIFF-UNet用于宫颈癌计划靶区的自动分割。该模型以编码器-解码器为主干结构，其中注意力分解卷积通过分解卷积核拓展感受野并提升计算效率；门控交互特征融合通过门控机制实现局部与全局特征的多尺度融合，增强模型对细节和上下文的综合表征；Transformer层用于捕捉长距离依赖进一步强化语义理解。选取陕西省肿瘤医院82例宫颈癌患者临床资料，选用Dice相似系数（DSC）、95%豪斯多夫距离（95HD）以及交并比（IOU）作为评估指标。结果显示，Trans-GIFF-UNet模型在测试集上的DSC为85.28%，IOU为74.61%，95HD为4.67 mm，性能优于UNet等对比模型，验证Trans-GIFF-UNet在宫颈癌计划靶区分割中的有效性与稳定性，可提高放疗医生的工作效率及勾画的一致性。
Abstract:
Abstract: Given that the target segmentation in radiotherapy planning is highly dependent on manual contouring, resulting in low efficiency and poor consistency, this study proposes a Trans-GIFF-UNet model based on the Transformer architecture for high-precision automatic segmentation of cervical cancer planning target volume. This model adopts an encoder-decoder as backbone structure. Inception separable convolution enlarges the receptive field and improves computational efficiency through kernel decomposition. Gated interactive feature fusion integrates multi-scale local and global features using a gating mechanism, enhancing the models representation of both fine details and contextual information. Transformer layers are incorporated to capture long-range dependencies and strengthen semantic understanding. Clinical data of 82 patients with cervical cancer were collected. Dice similarity coefficient, 95% Hausdorff distance, and intersection over union are used as evaluation metrics. Experimental results demonstrate that the Trans-GIFF-UNet model achieves a Dice similarity coefficient of 85.28%, an intersection over union of 74.61%, and a 95% Hausdorff distance of 4.67 mm on the test set, outperforming other comparative models including UNet. These findings validate the effectiveness and robustness of the Trans-GIFF-UNet in cervical cancer target segmentation, enhancing radiotherapy planning efficiency and segmentation consistency.

相似文献/References:

[1]万俊,聂生东,王远军,等.基于MRI的脑肿瘤分割技术研究进展[J].中国医学物理学杂志,2013,30(04):4266.[doi:10.3969/j.issn.1005-202X.2013.04.013]
[2]林江,戴齐,欧阳婷雪,等.一种边界和马尔可夫随机场相结合的脑MRI医学图像分割方法[J].中国医学物理学杂志,2015,32(05):717.[doi:doi:10.3969/j.issn.1005-202X.2015.05.023]
　[J].Chinese Journal of Medical Physics,2015,32(9):717.[doi:doi:10.3969/j.issn.1005-202X.2015.05.023]
[3]陶源,王佳飞,杜俊龙,等.基于卷积神经网络的细胞识别[J].中国医学物理学杂志,2017,34(1):53.[doi:10.3969/j.issn.1005-202X.2017.01.011]
　[J].Chinese Journal of Medical Physics,2017,34(9):53.[doi:10.3969/j.issn.1005-202X.2017.01.011]
[4]门阔,戴建荣. 利用深度反卷积神经网络自动勾画放疗危及器官[J].中国医学物理学杂志,2018,35(3):256.[doi:DOI:10.3969/j.issn.1005-202X.2018.03.002]
　MEN Kuo,DAI Jianrong. Automatic segmentation of organs at risk in radiotherapy using deep deconvolutional neural network[J].Chinese Journal of Medical Physics,2018,35(9):256.[doi:DOI:10.3969/j.issn.1005-202X.2018.03.002]
[5]邓金城,彭应林,刘常春,等. 深度卷积神经网络在放射治疗计划图像分割中的应用[J].中国医学物理学杂志,2018,35(6):621.[doi:DOI:10.3969/j.issn.1005-202X.2018.06.001]
　DENG Jincheng,PENG Yinglin,LIU Changchun,et al. Application of deep convolution neural network in radiotherapy planning image segmentation[J].Chinese Journal of Medical Physics,2018,35(9):621.[doi:DOI:10.3969/j.issn.1005-202X.2018.06.001]
[6]吴迪,胡胜,刘伟峰,等. 基于特征融合视觉显著性的医学图像分割[J].中国医学物理学杂志,2018,35(6):670.[doi:DOI:10.3969/j.issn.1005-202X.2018.06.010]
　WU Di,HU Sheng,LIU Weifeng,et al. Medical image segmentation based on visual saliency of feature fusion[J].Chinese Journal of Medical Physics,2018,35(9):670.[doi:DOI:10.3969/j.issn.1005-202X.2018.06.010]
[7]刘国才,官文静,田娟秀,等. 集成自适应回归核的肿瘤生物靶区随机游走勾画方法[J].中国医学物理学杂志,2018,35(7):758.[doi:DOI:10.3969/j.issn.1005-202X.2018.07.004]
　LIU Guocai,GUAN Wenjing,TIAN Juanxiu,et al. A random walk method with adaptive regression-kernel for delineation of biological target volumes of tumors[J].Chinese Journal of Medical Physics,2018,35(9):758.[doi:DOI:10.3969/j.issn.1005-202X.2018.07.004]
[8]查雪帆,杨丰,吴俣南,等. 结合迁移学习与深度卷积网络的心电分类研究[J].中国医学物理学杂志,2018,35(11):1307.[doi:DOI:10.3969/j.issn.1005-202X.2018.11.013]
　ZHA Xuefan,YANG Feng,WU Yunan,et al. ECG classification based on transfer learning and deep convolution neural network[J].Chinese Journal of Medical Physics,2018,35(9):1307.[doi:DOI:10.3969/j.issn.1005-202X.2018.11.013]
[9]安莹,黄能军,杨荣,等. 基于深度学习的心血管疾病风险预测模型[J].中国医学物理学杂志,2019,36(9):1103.[doi:DOI:10.3969/j.issn.1005-202X.2019.09.021]
　AN Ying,HUANG Nengjun,YANG Rong,et al. Deep learning-based model for risk prediction of cardiovascular diseases[J].Chinese Journal of Medical Physics,2019,36(9):1103.[doi:DOI:10.3969/j.issn.1005-202X.2019.09.021]
[10]徐航,随力,张靖雯,等.卷积神经网络在医学图像分割中的研究进展[J].中国医学物理学杂志,2019,36(11):1302.[doi:DOI:10.3969/j.issn.1005-202X.2019.11.011]
　XU Hang,SUI Li,ZHANG Jingwen,et al.Progress on convolutional neural network in medical image segmentation[J].Chinese Journal of Medical Physics,2019,36(9):1302.[doi:DOI:10.3969/j.issn.1005-202X.2019.11.011]
[11]宫进昌,赵尚义,王远军. 基于深度学习的医学图像分割研究进展[J].中国医学物理学杂志,2019,36(4):420.[doi:DOI:10.3969/j.issn.1005-202X.2019.04.010]
　GONG Jinchang,ZHAO Shangyi,WANG Yuanjun.Research progress on deep learning-based medical image segmentation[J].Chinese Journal of Medical Physics,2019,36(9):420.[doi:DOI:10.3969/j.issn.1005-202X.2019.04.010]
[12]张富利,杨安宁,路娜,等.基于FC_DenseNet深度学习网络自动分割肺癌放疗中的危及器官[J].中国医学物理学杂志,2021,38(2):259.[doi:DOI:10.3969/j.issn.1005-202X.2021.02.024]
　ZHANG Fuli,YANG Anning,LU Na,et al.FC_Densenet-based autosegmentation of organs-at-risk in lung cancer radiotherapy[J].Chinese Journal of Medical Physics,2021,38(9):259.[doi:DOI:10.3969/j.issn.1005-202X.2021.02.024]
[13]方新林,方艳红,王迪.基于多模态特征融合的脑瘤图像分割方法[J].中国医学物理学杂志,2022,39(6):682.[doi:DOI:10.3969/j.issn.1005-202X.2022.06.005]
　FANG Xinlin,FANG Yanhong,WANG Di.Brain tumor image segmentation method based on multi-modal feature fusion[J].Chinese Journal of Medical Physics,2022,39(9):682.[doi:DOI:10.3969/j.issn.1005-202X.2022.06.005]
[14]周金治,胡震,郭莉莉,等.基于GAN-DAUnet的肝脏CT图像肿瘤分割算法[J].中国医学物理学杂志,2023,40(8):971.[doi:DOI:10.3969/j.issn.1005-202X.2023.08.008]
　ZHOU Jinzhi,HU Zhen,et al.Liver CT image tumor segmentation algorithm based on GAN-DAUnet[J].Chinese Journal of Medical Physics,2023,40(9):971.[doi:DOI:10.3969/j.issn.1005-202X.2023.08.008]
[15]黄仟甲,张恒,李奇轩,等.医学图像分割的研究进展[J].中国医学物理学杂志,2024,41(8):939.[doi:DOI:10.3969/j.issn.1005-202X.2024.08.003]
　HUANG Qianjia,ZHANG Heng,,et al.Review on medical image segmentation methods[J].Chinese Journal of Medical Physics,2024,41(9):939.[doi:DOI:10.3969/j.issn.1005-202X.2024.08.003]

备注/Memo

备注/Memo:
【收稿日期】2026-04-21 【基金项目】国家重点研发计划（2016YFC0105103） 【作者简介】熊互湟，硕士，研究方向：辐射物理与医学物理，E-mail: 2509747580@qq.com 【通信作者】吴章文，研究员，研究方向：辐射物理与医学物理，E-mail: wuzhangwen@scu.edu.cn
更新日期/Last Update: 2026-09-30