胶质母细胞瘤放疗GTV自动分割的分区化评估
本研究回顾性分析100例术后胶质母细胞瘤(GBM)患者的多模态MRI,评估深度学习模型(Neosoma Glioma)按ESTRO-EANO指南自动分割GTV的临床适用性。最佳配置(切除腔+强化肿瘤外扩1 mm)平均Dice系数0.79,手动调整仅需5.9分钟(较手工勾画12.3分钟减少>50%),调整后自动分割与共识轮廓的剂量学等效。
要点
- 最佳自动分割配置为切除腔+强化肿瘤外扩1 mm,平均Dice 0.79。
- 手动调整自动分割仅需5.9分钟,较手工勾画12.3分钟减少>50%时间。
- 调整后自动分割的计划与共识轮廓计划剂量学等效,无临床相关差异。
核心结论
- Neosoma Glioma模型生成的术后GBM GTV分割具有临床实用性,几何性能与专家间变异相当,剂量学与共识轮廓等效,可减少>50%勾画时间。
研究设计与样本
- 回顾性分析2016-2020年Inselspital大学医院100例接受标准多模态MRI的GBM病例,比较自动分割GTV与专家轮廓的几何指标,并由放射肿瘤医生审核调整最佳配置,评估时间节省、几何相似性和剂量学影响。
主要发现
- 最佳自动分割(切除腔+强化肿瘤外扩1 mm)与真实轮廓的平均Dice系数为0.79(SD=0.14)。
- 手动调整自动分割耗时5.9(SD=4.6)分钟,手工勾画耗时12.3(SD=6.8)分钟,时间减少>50%。
- 自动分割与调整后GTV的平均Dice为0.84(SD=0.18)。
- 基于调整后自动分割的计划与基于共识轮廓的计划在靶区覆盖和危及器官保护方面无临床相关差异。
关键数值与推荐
- 最佳自动分割配置为切除腔加强化肿瘤外扩1 mm。
- 自动分割与真实轮廓的平均Dice系数为0.79(SD=0.14)。
- 手动调整自动分割耗时5.9(SD=4.6)分钟,手工勾画耗时12.3(SD=6.8)分钟。
- 自动分割与调整后GTV的平均Dice为0.84(SD=0.18)。
临床怎么用
- 可将Neosoma Glioma模型用于术后GBM的GTV自动分割,经放射肿瘤医生手动调整后用于放疗计划,以缩短勾画时间并标准化靶区定义。
局限性
- 本研究为单中心回顾性分析,样本量100例,需外部验证。
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
BACKGROUND AND PURPOSE: Manual delineation of target volumes in glioblastoma (GBM) radiotherapy (RT) is time-consuming and variable. This study evaluates the clinical applicability of a preliminary deep learning model (Neosoma Glioma) for automating gross tumor volume (GTV) segmentation in postoperative GBM per ESTRO-EANO guidelines. MATERIALS AND METHODS: We retrospectively analyzed 100 GBM cases treated at Inselspital University Hospital, Bern (2016-2020) with standardized multi-modal MRI. Auto-segmented GTVs were compared to expert-defined contours using geometric metrics. Radiation oncologists reviewed and adjusted the best-performing configuration. Time savings, geometric similarity, and dosimetric impact were assessed. RESULTS: Optimal auto-segmentation (resection cavity plus enhancing tumor with 1 mm margin) achieved a mean Dice similarity coefficient of 0.79 (SD = 0.14) vs. ground truth. Manual adjustment took 5.9 (SD = 4.6) minutes vs. 12.3 (SD = 6.8) minutes for manual contouring (>50 % time reduction). The mean Dice between auto-segmented and adjusted GTVs was 0.84 (SD = 0.18). Dosimetric evaluation showed plans from adjusted auto-segmentations were equivalent to those based on consensus contours, with no clinically relevant differences in target coverage or organ-at-risk sparing. CONCLUSION: The Neosoma Glioma model generates clinically useful postoperative GTV segmentations, with geometric performance comparable to expert variability and dosimetric equivalence to consensus contours. It reduces contouring time by over 50%, enabling faster RT workflows. Its consistency across diverse GBM presentations supports its practical value. AI-based segmentation can help standardize GBM target definition when integrated into RT planning with proper quality assurance.
引用
- Poel R, Mose L, Reinhardt P, et al. Evaluation of compartmentalized automatic segmentation for definition of the GTV in glioblastoma radiotherapy. Radiother Oncol. 2025. PMID: 41308924. doi:10.1016/j.radonc.2025.111308
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。