王妍灵,梁 永

乳腺癌术后放疗是提高局部控制率的核心手段。靶区勾画作为其关键环节，效率与质量受传统手动勾画局限制约。基于深度学习的人工智能（AI）技术为突破该瓶颈提供了颠覆性解决方案。本文系统梳理该领域智能勾画技术的应用现状及发展趋势：在高对比度危及器官（OAR）已达临床应用水平，显著提升效率与一致性；在术后解剖复杂、边界模糊的临床靶区（CTV）及淋巴引流区（RLN）虽获进展，仍面临模型泛化能力不足与微小病灶识别困难的挑战。当前，数据孤岛、模型可解释性差及与临床工作流整合难题制约其广泛应用。未来，联邦学习、多模态融合及可解释AI将是推动技术从“可用”迈向“可靠”的核心驱动力。

乳腺恶性肿瘤；放射疗法；靶区勾画

[1] 王裕新,潘凯枫,李文庆.2022全球癌症统计报告解读[J].肿瘤综合治疗电子杂志,2024,10(3):1-16. [2] 王少明,郑荣寿,韩冰峰,等.2022年中国人群恶性肿瘤发病与死亡年龄特征分析[J].中国肿瘤,2024,33(03):165-174. [3] Buelens P,Willems S,Vandewinckele L,et al.Clinical evaluation of a deep learning model for segmentation of target volumes in breast cancer radiotherapy[J].Radiother Oncol,2022,171:84-90. [4] Kim Y W,Biggs S,Claridge Mackonis E.Investigation on performance of multiple AI-based auto-contouring systems in organs at risks(OARs)delineation[J].Physical and Engineering Sciences in Medicine,2024,47(3):1123-1140. [5] Chung S Y,Chang J S,Choi M S,et al.Clinical feasibility of deep learning-based auto-segmentation of target volumes and organs-at-risk in breast cancer patients after breast-conserving surgery[J].Radiation Oncology,2021,16(1):44. [6] 许卓华,杨慧,江舟,等.基于3D ResSE-Unet的智能靶区勾画在乳腺癌术后放疗中的应用研究[J].中华放射医学与防护杂志,2023, 43(4):269-275. [7] Chen J,Mei J,Li X,et al.TransUNet:Rethinking the U-Net architecture design for medical image segmentation through the lens of transformers[J].Med Image Anal,2024,97:103280. [8] 刘方杰.基于卷积神经网络的乳腺癌放射治疗临床靶区及危及器官自动勾画[D].北京协和医学院,2020. [9] Zhao C,Qin B,Feng S,et al.Hyperspectral Image Classification With Multi-Attention Transformer and Adaptive Superpixel Segmentation-Based Active Learning[J].IEEE Trans Image Process,2023,32:3606-3621. [10] Zhang X,Sun G,Jia X,et al.Spectral–Spatial Self-Attention Networks for Hyperspectral Image Classification[J].IEEE Transactions on Geoscience and Remote Sensing,2022,60({}):1-15. [11] Anari S,de Oliveira G G,Ranjbarzadeh R,et al.EfficientUNetViT:Efficient Breast Tumor Segmentation Utilizing UNet Architecture and Pretrained Vision Transformer[J].Bioengineering(Basel),2024,11(9):945. [12] Guan S,Loew M.Breast cancer detection using synthetic mammograms from generative adversarial networks in convolutional neural networks[J].J Med Imaging(Bellingham),2019,6(3):31411. [13] Pang T,Wong J,Ng W L,et al.Semi-supervised GAN-based Radiomics Model for Data Augmentation in Breast Ultrasound Mass Classification[J].Comput Methods Programs Biomed,2021,203:106018. [14] Kang S K,An H J,Jin H,et al.Synthetic CT generation from weakly paired MR images using cycle-consistent GAN for MR-guided radiotherapy[J].Biomed Eng Lett,2021,11(3):263-271. [15] Zhao Y,Wang H,Yu C,et al.Compensation cycle consistent generative adversarial networks(Comp-GAN)for synthetic CT generation from MR scans with truncated anatomy[J].Med Phys,2023,50(7):4399-4414. [16] 李成梁.基于深度学习的乳腺癌放疗靶区勾画和剂量预测[D].山东大学,2024. [17] Ronneberger O,Fischer P,Brox T.U-Net:Convolutional Networks for Biomedical Image Segmentation,Cham,2015[C].Springer International Publishing,2015-01-01. [18] Zhou Z,Siddiquee M,Tajbakhsh N,et al.UNet++:Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation[J].IEEE Trans Med Imaging,2020,39(6):1856-1867. [19] Oktay O,Schlemper J,Le Folgoc L I C,et al.Attention U-Net:Learning Where to Look for the Pancreas[J].ArXiv,2018,abs/1804.03999. [20] 韩笑辰,韩若凌,迟子锋,等.乳腺癌RIHD心脏亚结构模型及人工智能研究现状[J].河北医科大学学报,2024,45(07):861-865. [21] Emin S,Rossi E,Hedman M,et al.Performance of multi-vendor auto-segmentation models for thoracic organs at risk trained on a single dataset[J].Phys Med,2025,137:105089. [22] Sun R,Li X,Han B,et al.Multi-task learning for joint prediction of breast cancer histological indicators in dynamic contrast-enhanced magnetic resonance imaging[J].Comput Methods Programs Biomed,2025,267:108830. [23] 龚筱钦.基于深度学习软件的早期乳腺癌靶区及危及器官自动勾画技术在临床的应用研究[D].苏州大学,2022. [24] Doolan P J,Charalambous S,Roussakis Y,et al.A clinical evaluation of the performance of five commercial artificial intelligence contouring systems for radiotherapy[J].Front Oncol,2023,13:1213068. [25] Byun H K,Chang J S,Choi M S,et al.Evaluation of deep learning-based autosegmentation in breast cancer radiotherapy[J].Radiat Oncol,2021,16(1):203. [26] Liu Z,Liu F,Chen W,et al.Automatic Segmentation of Clinical Target Volumes for Post-Modified Radical Mastectomy Radiotherapy Using Convolutional Neural Networks[J].Front Oncol,2020,10:581347. [27] Men K,Zhang T,Chen X,et al.Fully automatic and robust segmentation of the clinical target volume for radiotherapy of breast cancer using big data and deep learning[J].Phys Med,2018,50:13-19. [28] Xu Q,Xi X,Meng X,et al.Difficulty-aware bi-network with spatial attention constrained graph for axillary lymph node segmentation[J]. Science China Information Sciences,2022,65(9):192102. [29] Qiao X,Jiang C,Li P,et al.Improving Breast Tumor Segmentation in PET via Attentive Transformation Based Normalization[J].IEEE J Biomed Health Inform,2022,26(7):3261-3271. [30] Barragán-Montero A,Bibal A,Dastarac M H,et al.Towards a safe and efficient clinical implementation of machine learning in radiation oncology by exploring model interpretability,explainability and data-model dependency[J].Phys Med Biol,2022,67(11). [31] Lin H,Xiao H,Dong L,et al.Deep learning for automatic target volume segmentation in radiation therapy:a review[J].Quant Imaging Med Surg,2021,11(12):4847-4858. [32] Cardenas C E,Yang J,Anderson B M,et al.Advances in Auto-Segmentation[J].Semin Radiat Oncol,2019,29(3):185-197. [33] Gu R,Wang G,Song T,et al.CA-Net:Comprehensive Attention Convolutional Neural Networks for Explainable Medical Image Segmentation[J].IEEE Trans Med Imaging,2021,40(2):699-711. [34] Groen A M,Kraan R,Amirkhan S F,et al.A systematic review on the use of explainability in deep learning systems for computer aided diagnosis in radiology:Limited use of explainable AI?[J].Eur J Radiol,2022,157:110592. [35] L.Z,X.W,D.Y,et al.Generalizing Deep Learning for Medical Image Segmentation to Unseen Domains via Deep Stacked Transformation[J].IEEE Transactions on Medical Imaging,2020,39(7):2531-2540. [36] Alami H,Lehoux P,Papoutsi C,et al.Understanding the integration of artificial intelligence in healthcare organisations and systems through the NASSS framework:a qualitative study in a leading Canadian academic centre[J].BMC Health Serv Res,2024,24(1):701. [37] L.C,T.R,E.V,et al.Federated Learning Applications for Breast Cancer:2023 IEEE International Conference on Bioinformatics and Biomedicine(BIBM),2023[C].20235-8 Dec.2023. [38] Shukla S,Rajkumar S,Sinha A,et al.Federated learning with differential privacy for breast cancer diagnosis enabling secure data sharing and model integrity[J].Sci Rep,2025,15(1):13061. [39] Wang Z,Cao N,Sun J,et al.Uncertainty estimation-and attention-based semi-supervised models for automatically delineate clinical target volume in CBCT images of breast cancer[J].Radiation Oncology,2024,19(1):66. [40] Wang H,Huai L,Li W,et al.WeakMedSAM:Weakly-Supervised Medical Image Segmentation via SAM With Sub-Class Exploration and Prompt Affinity Mining[J].IEEE Trans Med Imaging,2025,44(7):2795-2807. [41] Chen Y,Yang X H,Wei Z,et al.Generative Adversarial Networks in Medical Image augmentation:A review[J].Comput Biol Med,2022,144:105382. [42] Zhai Y,Chen Z,Shao D.[Breast cancer lesion segmentation based on co-learning feature fusion and Transformer][J].Sheng Wu Yi Xue Gong Cheng Xue Za Zhi,2024,41(2):237-245. [43] Talaat F M,Gamel S A,El-Balka R M,et al.Grad-CAM Enabled Breast Cancer Classification with a 3D Inception-ResNet V2:Empowering Radiologists with Explainable Insights[J].Cancers(Basel),2024,16(21). [44] Smith A G,Petersen J,Terrones-Campos C,et al.RootPainter3D:Interactive-machine-learning enables rapid and accurate contouring for radiotherapy[J].Med Phys,2022,49(1):461-473.

点击下载PDF