| 9 | 0 | 25 |
| 下载次数 | 被引频次 | 阅读次数 |
目的 面向在线自适应放疗中头颈部危及器官自动勾画的快速审核需求,以剂量学指标为基准,评估几何相似性指标的适用性,并构建兼顾漏检风险与审核效率的快速筛查模型。方法 回顾性纳入29例头颈部放疗患者,获得243对危及器官自动/人工勾画轮廓;计算自动/人工勾画轮廓之间的几何相似性指标,如戴斯相似性指数(Dice)、豪斯多夫距离(HD95/100)等。同时基于临床计划剂量分布计算自动/人工勾画轮廓下剂量学指标差异,如平均/最大剂量等,分析几何相似性指标与剂量学差异的相关性,最后以3 Gy为阈值,定义剂量学不一致事件Y3Gy。以各几何指标为自变量,训练逻辑斯蒂回归模型预估Y3Gy发生概率,通过受试者工作特征曲线与精确率-召回率曲线及曲线下面积评估分类准确性,并结合阈值扫描确定最优阈值区间。结果 本工作中所考虑的几何相似性与剂量学指标差异存在弱到中等相关性,相比最大剂量差,几何相似性与平均剂量差的相关性更为稳定;在己调查的多项几何指标中,HD95对Y3Gy的分类能力最优,阈值扫描提示将阈值设置于在4～9 mm之间,可兼顾减少漏检风险与提升自动勾画质控效率。结论 本研究中调查的部分几何相似性指标与剂量学指标差异存在弱到中等相关性;在本研究所调查的头颈部危及器官勾画质控任务中,HD95表现出最好的预测精度,可快速完成自动勾画质量控制,有望提升自适应放疗效率。
Abstract：Objective To address the need for rapid review of automated organ-at-risk(OAR) contouring for head-andneck cancer in online adaptive radiotherapy,this study used dosimetric indices as the reference standard to evaluate the suitability of geometric similarity metrics and to develop a rapid screening model that balances the risk of missed errors with review efficiency.Methods This retrospective study included 29 patients with head-and-neck radiotherapy,yielding 243pairs of OAR contours(auto-generated vs.manual).Geometric similarity metrics,including the Dice similarity cofficient(Dice),the 95th percentile Hausdorff distance(HD95),and the maximum Hausdorff distance(HD 100),were computed between the two contour sets.Using the dose distributions from clinical treatment plans,dosimetric differences between the two contour sets were calculated for key indices,including mean dose(Dmean) and maximum dose(Dmax).Correlations between geometric metrics and dosimetric differences were analyzed.Dosimetric discrepancy events were defined as Y3Gyusing a threshold of 3 Gy.Within a univariable logistic regression triage framework,each geometric metric was used as an independent variable to estimate the probability of Y3Gy.Discriminative performance was assessed using receiver operating characteristic(ROC) curves and precision-recall(PR) curves,with the area under the curve(AUC) including ROC-AUC and PR-AUC.Practical operating thresholds were determined via threshold sweeping.Results Geometric similarity metrics showed weak-to-moderate correlations with dosimetric differences;compared with maximum dose,correlations with mean dose differences were more consistent.Among the evaluated metrics,HD95 achieved the best classification performance for Y3Gy.Threshold sweeping suggested that an HD95 threshold in the range of 4-9 mm can balance the risk of missed discrepancies with the efficiency of automated contour quality assurance(QA).Conclusion Several geometric similarity metrics demonstrated only weak-to-moderate associations with dosimetric differences.For head-and-neck OAR contour QA in this study,HD95 provided the best discriminative performance and can support rapid triage of automated contours,with a practical operating threshold range of 4-9 mm,potentially improving the efficiency of adaptive radiotherapy workflows.
[1]Brodin NP, ToméWA. Revisiting the dose constraints for head and neck OARs in the current era of IMRT[J]. Oral Oncol, 2018, 86:8-18. DOI:10.1016/j.oraloncology.2018.08.018.
[2]Matoska T, Patel M, Liu HF, et al. Review of deep learning based autosegmentation for clinical target volume:current status and future directions[J]. Adv Radiat Oncol, 2024, 9(5):101470. DOI:10.1016/j.adro.2024.101470.
[3]Bentzen SM, Constine LS, Deasy JO, et al. Quantitative analyses of normal tissue effects in the clinic(QUANTEC):an introduction to the scientific issues[J]. Int J Radiat Oncol Biol Phys, 2010,76(S3):S3-S9. DOI:10.1016/j.ijrobp.2009.09.040.
[4]De Biase A, Sijtsema NM, Janssen T, et al. Clinical adoption of deep learning target auto-segmentation for radiation therapy:challenges, clinical risks, and mitigation strategies[J]. BJR|Artif Intell, 2024, 1(1):ubae015. DOI:10.1093/bjrai/ubae015.
[5]Sherer MV, Lin D, Elguindi S, et al. Metrics to evaluate the performance of auto-segmentation for radiation treatment planning:a critical review[J]. Radiother Oncol, 2021, 160:185-191. DOI:10.1016/j.radonc.2021.05.003.
[6]Pang EPP, Tan HQ, Wang FQ, et al. Multicentre evaluation of deep learning CT autosegmentation of the head and neck region for radiotherapy[J]. npj Digit Med, 2025, 8(1):312. DOI:10.1038/s41746-025-01624-z.
[7]张汉东,李晓辉,殷鲁旭,等.深度学习驱动骨肉瘤智能放疗的最新进展与挑战[J].中国辐射卫生，2025,34(6):912-917. DOI:10.13491/j.issn.1004-714X.2025.06.021.Zhang HD, Li XH, Yin LX, et al. Deep learning-driven intelligent radiotherapy for osteosarcoma:recent advances and challenges[J].Chin J Radiol Health, 2025, 34(6):912-917. DOI:10.13491/j.issn.1004-714X.2025.06.021.(in Chinese)
[8]Vandewinckele L, Claessens M, Dinkla A, et al. Overview of artificial intelligence-based applications in radiotherapy:recommendations for implementation and quality assurance[J]. Radiother Oncol, 2020, 153:55-66. DOI:10.1016/j.radonc.2020.09.008.
[9]Chlap P, Min H, Martin J, et al. Implementation of an automated contour quality assurance tool within the TROG 18.01 NINJA trial[J]. Radiother Oncol, 2026, 214:111269. DOI:10.1016/j.radonc.2025.111269.
[10]Mackay K, Bernstein D, Glocker B, et al. A review of the metrics used to assess auto-contouring systems in radiotherapy[J]. Clin Oncol, 2023, 35(6):354-369. DOI:10.1016/j.clon.2023.01.016.
[11]Le Bao V, Haworth A, Dowling J, et al. Evaluating the relationship between contouring variability and modelled treatment outcome for prostate bed radiotherapy[J]. Phys Med Biol, 2024, 69(8):085008. DOI:10.1088/1361-6560/ad3325.
[12]Marquez B, Wooten ZT, Salazar RM, et al. Analyzing the relationship between dose and geometric agreement metrics for auto-contouring in head and neck normal tissues[J]. Diagnostics(Basel), 2024, 14(15):1632. DOI:10.3390/diagnostics14151632.
[13]Xian LX, Li GJ, Xiao Q, et al. Clinically oriented target contour evaluation using geometric and dosimetric indices based on simple geometric transformations[J]. Technol Cancer Res Treat, 2021, 20:15330338211036325. DOI:10.1177/15330338211036325.
[14]Min H, Dowling J, Jameson MG, et al. Clinical target volume delineation quality assurance for MRI-guided prostate radiotherapy using deep learning with uncertainty estimation[J]. Radiother Oncol, 2023, 186:109794. DOI:10.1016/j.radonc.2023.109794.
[15]Liu ZK, Liu X, Guan H, et al. Development and validation of a deep learning algorithm for auto-delineation of clinical target volume and organs at risk in cervical cancer radiotherapy[J].Radiother Oncol, 2020, 153:172-179. DOI:10.1016/j.radonc.2020.09.060.
[16]Brouwer CL, Steenbakkers RJHM, Bourhis J, et al. CT-based delineation of organs at risk in the head and neck region:DAHANCA, EORTC, GORTEC, HKNPCSG, NCIC CTG, NCRI,NRG Oncology and TROG consensus guidelines[J]. Radiother Oncol, 2015, 117(1):83-90. DOI:10.1016/j.radonc.2015.07.041.
[17]Rüfenacht E, Kamath A, Suter Y, et al. PyRaDiSe:a Python package for DICOM-RT-based auto-segmentation pipeline construction and DICOM-RT data conversion[J]. Comput Methods Programs Biomed, 2023, 231:107374. DOI:10.1016/j.cmpb.2023.107374.
[18]van der Veen J, Willems S, Deschuymer S, et al. Benefits of deep learning for delineation of organs at risk in head and neck cancer[J]. Radiother Oncol, 2019, 138:68-74. DOI:10.1016/j.radonc.2019.05.010.
[19]Tambas M, Steenbakkers RJHM, van der Laan HP, et al. First experience with model-based selection of head and neck cancer patients for proton therapy[J]. Radiother Oncol, 2020, 151:206-213. DOI:10.1016/j.radonc.2020.07.056.
[20]Lim TY, Gillespie E, Murphy J, et al. Clinically oriented contour evaluation using dosimetric indices generated from automated knowledge-based planning[J]. Int J Radiat Oncol Biol Phys, 2019,103(5):1251-1260. DOI:10.1016/j.ijrobp.2018.11.048.
[21]van Rooij W, Dahele M, Brandao HR, et al. Deep learning-based delineation of head and neck organs at risk:geometric and dosimetric evaluation[J]. Int J Radiat Oncol Biol Phys, 2019, 104(3):677-684. DOI:10.1016/j.ijrobp.2019.02.040.
[22]Kaderka R, Gillespie EF, Mundt RC, et al. Geometric and dosimetric evaluation of atlas based auto-segmentation of cardiac structures in breast cancer patients[J]. Radiother Oncol, 2019, 131:215-220. DOI:10.1016/j.radonc.2018.07.013.
基本信息:
DOI：10.13491/j.issn.1004-714X.2026.02.006
中图分类号:R730.55
引用信息:
[1]杨晓雨,姚凯宁,蒲亦晨,等.头颈部危及器官自动勾画的快速质量控制方法[J].中国辐射卫生,2026,35(02):193-199.DOI:10.13491/j.issn.1004-714X.2026.02.006.
基金信息:
国家重大研发计划项目（2019YFF01014405,2023YFF0613501）; 北京大学肿瘤医院科学研究基金（ZY202410）; 核药研发转化与精准防护山西省重点实验室开放基金（TNMPP-2025-02）; 国家自然科学基金（12375335）; 北京市自然科学基金（1202009,25JL001）
2026-04-15
2026-04-15