不同自动分割模型用于乳腺癌术后放疗及区域淋巴结照射靶区勾画的验证
本研究将三种基于人工智能和ESTRO指南的商业分割模型(M1-3)应用于50例连续接受乳腺癌术后局部放疗(含区域淋巴结照射)的参考患者,勾画临床靶区。结果显示,无需或仅需轻微修正的病例比例分别为72%(M1)、64%(M2)和78%(M3);计划靶区的中位DICE值为0.87-0.88,Hausdorff值为2.96-3.55;临床可用性与DICE指数显著相关,界值为0.82-0.86。
要点
- 三种AI分割模型在乳腺癌术后放疗靶区勾画中无需或仅需轻微修正的比例为72%(M1)、64%(M2)和78%(M3)。
- 计划靶区中位DICE值为0.87-0.88,中位Hausdorff值为2.96-3.55。
- 临床可用性与DICE指数显著相关,无需或轻微调整的DICE界值为0.82-0.86。
核心结论
- 基于人工智能的自动分割程序显示出高准确度,为基于指南的靶区勾画提供了标准化和高效支持,是全自动放疗计划工作流程的前提。
研究设计与样本
- 本研究为验证性研究,将三种商业可用的、基于人工智能和ESTRO指南的分割模型(M1-3)应用于50例连续接受乳腺癌术后局部放疗(含区域淋巴结照射)的参考患者,勾画临床靶区。
主要发现
- 分割模型生成的几何结构与每位患者的参考体积比较,无需或仅需轻微修正的病例比例分别为72%(M1)、64%(M2)和78%(M3)。
- 计划靶区的中位DICE值为0.87-0.88,中位Hausdorff值为2.96-3.55。
- 临床可用性与DICE指数显著相关,计算出的界值为0.82-0.86,用于定义无需或轻微调整。
- 右侧或左侧靶区以及呼吸方法(深吸气屏气 vs. 自由呼吸)不影响生成结构的质量。
关键数值与推荐
- 无需或仅需轻微修正的病例比例:M1为72%,M2为64%,M3为78%。
- 计划靶区中位DICE值:0.87-0.88;中位Hausdorff值:2.96-3.55。
- 临床可用性与DICE指数显著相关,无需或轻微调整的DICE界值为0.82-0.86。
临床怎么用
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
INTRODUCTION: Target volume delineation is routinely performed in postoperative radiotherapy (RT) for breast cancer patients, but it is a time-consuming process. The aim of the present study was to validate the quality, clinical usability and institutional-specific implementation of different auto-segmentation tools into clinical routine. METHODS: Three different commercially available, artificial intelligence-, ESTRO-guideline-based segmentation models (M1-3) were applied to fifty consecutive reference patients who received postoperative local RT including regional nodal irradiation for breast cancer for the delineation of clinical target volumes: the residual breast, implant or chestwall, axilla levels 1 and 2, the infra- and supraclavicular regions, the interpectoral and internal mammary nodes. Objective evaluation metrics of the created structures were conducted with the Dice similarity index (DICE) and the Hausdorff distance, and a manual evaluation of usability. RESULTS: The resulting geometries of the segmentation models were compared to the reference volumes for each patient and required no or only minor corrections in 72 % (M1), 64 % (M2) and 78 % (M3) of the cases. The median DICE and Hausdorff values for the resulting planning target volumes were 0.87-0.88 and 2.96-3.55, respectively. Clinical usability was significantly correlated with the DICE index, with calculated cut-off values used to define no or minor adjustments of 0.82-0.86. Right or left sided target and breathing method (deep inspiration breath hold vs. free breathing) did not impact the quality of the resulting structures. CONCLUSION: Artificial intelligence-based auto-segmentation programs showed high-quality accuracy and provided standardization and efficient support for guideline-based target volume contouring as a precondition for fully automated workflows in radiotherapy treatment planning.
引用
- Meixner E, Glogauer B, Klüter S, et al. Validation of different automated segmentation models for target volume contouring in postoperative radiotherapy for breast cancer and regional nodal irradiation. Clin Transl Radiat Oncol. 2024. PMID: 39308634. doi:10.1016/j.ctro.2024.100855
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。