人工智能技术在乳腺癌放疗靶区勾画中的应用研究-国际临床研究杂志-CSCIED科技核心评价数据库-手机版

人工智能技术在乳腺癌放疗靶区勾画中的应用研究

Application of three-dimensional depth-convolution neural network in clinic target volume of post-mastectomy radiation therapy

ES评分 0 浏览量：740 下载量：0

全文阅读下载引用收藏

DOI	10.12208/j.ijcr.20220256
刊名	国际临床研究杂志 International Journal of Clinical Research
年，卷(期)	2022, 6(6)
作者	江舟,黄海欣,杨慧,陆颖*,
作者单位	广西柳州市工人医院（广西医科大学第四附属医院）肿瘤科广西柳州 ;
摘要	评估基于深度学习人工智能技术的放疗靶区自动勾画软件在临床应用中的价值。方法收集2018年9月-2022年5月在广西柳州市工人医院肿瘤科行乳腺癌乳房切除术后辅助放疗的380例患者纳入研究。将U-Net自动勾画靶区，与高年资医生手动勾画靶区进行对比，计算戴斯相似性系数（Dice Similarityco Efficient，DSC）和豪斯多夫距离（95% Hausdorff Distance, 95%HD）。结果胸壁野的DSC，左侧为0.851±0.036，右侧为0.834±0.044；95%HD，左侧为6.579±2.890mm，右侧为9.250±7.811mm。锁骨上野的DSC，左侧为0.806±0.051，右侧为0.823±0.062；95% HD，左侧为6.823±2.695mm，右侧为6.468±2.996mm。结论基于深度学习人工智能技术的自动勾画靶区与人工勾画相比有较好的相似性和一致性，该技术有助于提高乳腺癌放射治疗靶区的规范性及临床效率。
Abstract	Objective To evaluate the value of automatic delineation software of radiotherapy target area based on deep learning artificial intelligence technology in clinical application. Methods 380 patients who underwent adjuvant radiotherapy after mastectomy for breast cancer in the Department of Oncology, Liuzhou Workers Hospital, Guangxi from September 2018 to August 2021 were collected and included in the study. After automatic delineation of the target area using U-Net, manual modification was completed by senior physicians, and the target area before and after modification was compared to calculate the Dice Similarityco Efficient (DSC) and 95%Hausdorff Distance (95%HD). Results The DSC of chest wall field was 0.851±0.036 on the left side and 0.834±0.044 on the right side; 95%HD was 6.579±2.890mm on the left side and 9.250±7.811mm on the right side. The DSC of the supraclavicular field was 0.806±0.051 on the left and 0.823±0.062 on the right; 95%HD was 6.823±2.695mm on the left and 6.468±2.996mm on the right. Conclusion The automatic delineation software based on deep learning artificial intelligence technology has a high accuracy in delineating the target area of postoperative radiotherapy for breast cancer and can be used in clinical work with slight modification.
关键词	乳腺癌；靶区；自动勾画；放疗
KeyWord	Breast Cancer; Clinic Target Volume; Automatic Delineation; Radiotherapy
基金项目
页码	52-59

参考文献
相关文献

[1] Velker V M, Rodrigues G B, Dinniwell R, et al. Creation of RTOG compliant patient CT-atlases for automated atlas based contouring of local regional breast and high-risk prostate cancers[J]. Radiation Oncology, 2013, 8(1):188.
[2] Barley S, Antoine C, Webster G, et al. Atlas-based Auto-contouring – Balancing Accuracy with Efficiency in OnQ rts[J]. European Oncology and Haematology, 2014, 10(2).
[3] 张玉海, 李月敏. 人工智能在肿瘤放射治疗中的研究进展[J]. 实用肿瘤学杂志, 2019, 33(6):5.
[4] U-Net: deep learning for cell counting, detection, and morphometry[J]. Nature Methods, 2019.
[5] Zunair H, Hamza A B. Sharp U-Net: Depthwise Convolutional Network for Biomedical Image Segmentation[J]. Computers in Biology and Medicine, 2021.
[6] Taha A A , Hanbury A . Metrics for evaluating 3D medical image segmentation: analysis, selection, and tool[J]. BMC Medical Imaging,15,1(2015-08-12), 2015, 15(29).
[7] Long, Jonathan, Shelhamer, et al. Fully Convolutional Networks for Semantic Segmentation[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017.
[8] Bridge P, Bridge R. Artificial Intelligence in Radiotherapy: A Philosophical Perspective[J]. Journal of Medical Imaging and Radiation Sciences, 2019, 50(4S2).
[9] Offersen B V, Boersma L J, Kirkove C, et al. ESTRO consensus guideline on target volume delineation for elective radiation therapy of early stage breast cancer, version 1.1[J]. Radiotherapy and Oncology, 2016.
[10] Likhacheva A. Contouring Guidelines for the Axillary Lymph Nodes for the Delivery of Radiation Therapy in Breast Cancer: Evaluation of the RTOG Breast Cancer Atlas: Gentile MS, Usman AA, Neuschler EI, etal (Robert H. Lurie Comprehensive Cancer Ctr of Northwestern Univ, [J]. Breast Diseases: A Year Book Quarterly, 2016, 27( 2):156-157.
[11] Wong J , Fong A , Mcvicar N , et al. Comparing deep learning-based auto-segmentation of organs at risk and clinical target volumes to expert inter-observer variability in radiotherapy planning[J]. Radiotherapy and Oncology, 2019, 144:152-158.
[12] Internal and external validation of an ESTRO delineation guideline - dependent automated segmentation tool for loco-regional radiation therapy of early breast cancer[J]. Radiotherapy and oncology: Journal of the European Society for Therapeutic Radiology and Oncology, 2016, 121(3):424-430.
[13] Choi MS, Choi BS, Chung SY, et al. Clinical evaluation of atlas- and deep learning-based automatic segmentation of multiple organs and clinical target volumes for breast cancer[J]. Radiother Oncol. 2020;153:139-145.
[14] Kiljunen T , Akram S , Niemel J , et al. A Deep Learning-Based Automated CT Segmentation of Prostate Cancer Anatomy for Radiation Therapy Planning-A Retrospective Multicenter Study[J]. Diagnostics, 2020, 10(11):959.
[15] Brunenberg E , Steinseifer I K , Bosch S , et al. External validation of deep learning-based contouring of head and neck organs at risk[J]. Physics and Imaging in Radiation Oncology, 2020, 15:8-15.
[16] Pmpp A , St B , Gnmc D , et al. Winter is over: The use of Artificial Intelligence to individualise radiation therapy for breast cancer[J]. The Breast, 2020, 49:194-200.
[17] 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]. Physica Medica, 2018, 50:13-19.
[18] Chen X , Men K , Chen B , et al. CNN-Based Quality Assurance for Automatic Segmentation of Breast Cancer in Radiotherapy[J]. Frontiers in Oncology, 2020, 10:524.

引用本文

江舟,黄海欣,杨慧,陆颖*. 人工智能技术在乳腺癌放疗靶区勾画中的应用研究 [J]. 国际临床研究杂志. 2022; 6; (6). 52 - 59.

文献评论

相关学者

相关机构