基于深度学习的宫颈癌大体肿瘤体积磁共振图像近距离放疗自动分割
本研究回顾性纳入195例宫颈癌患者共524个近距离放疗分次,采用3D nnU-Net在轴位T2加权MRI上自动分割大体肿瘤体积(GTV),并与临床手工分割比较。中位Dice为0.73,中位D90和D98差异分别为0.18 Gy和0.20 Gy;不同FIGO分期之间几何与剂量学性能无显著差异,较大肿瘤的Dice和剂量学性能显著更优。
要点
- 3D nnU-Net自动分割宫颈癌近距离放疗MRI的GTV,中位Dice为0.73。
- 自动与手工分割的中位ΔD90为0.18 Gy,中位ΔD98为0.20 Gy。
- 不同FIGO分期之间几何与剂量学性能无显著差异。
核心结论
- nnU-Net框架在近距离放疗MRI图像上分割宫颈癌GTV达到当前最优性能,几何和剂量学中位性能合理但患者间变异性高。
研究设计与样本
- 回顾性收集2012年8月至2021年12月间195例宫颈癌患者,共纳入524个独立近距离放疗分次,使用轴位T2加权MRI序列。
- 采用3D nnU-Net作为自动分割框架,将自动分割与临床实践使用的手工分割进行比较。
- 以Sørensen-Dice系数、95百分位Hausdorff距离和平均表面距离评估几何性能,以临床剂量分布下D98和D90的差值评估剂量学影响。
- 还分别按FIGO分期和GTV体积比较网络性能。
主要发现
- 网络获得中位Dice为0.73(四分位距=0.50-0.80),中位95百分位Hausdorff距离为6.8 mm(四分位距=4.2-12.5 mm),中位平均表面距离为1.4 mm(四分位距=0.90-2.8 mm)。
- 中位ΔD90为0.18 Gy(四分位距=-1.38-1.19 Gy),中位ΔD98为0.20 Gy(四分位距=-1.10-0.95 Gy)。
- 不同FIGO分期的肿瘤之间未观察到几何或剂量学性能的显著差异,但较大肿瘤的Dice和剂量学性能显著改善。
临床怎么用
- 原文未据此给出临床推荐。
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
BACKGROUND: Segmentation of the Gross Tumor Volume (GTV) is a crucial step in the brachytherapy (BT) treatment planning workflow. Currently, radiation oncologists segment the GTV manually, which is time-consuming. The time pressure is particularly critical for BT because during the segmentation process the patient waits immobilized in bed with the applicator in place. Automatic segmentation algorithms can potentially reduce both the clinical workload and the patient burden. Although deep learning based automatic segmentation algorithms have been extensively developed for organs at risk, automatic segmentation of the targets is less common. The aim of this study was to automatically segment the cervical cancer GTV on BT MRI images using a state-of-the-art automatic segmentation framework and assess its performance. METHODS: A cohort of 195 cervical cancer patients treated between August 2012 and December 2021 was retrospectively collected. A total of 524 separate BT fractions were included and the axial T2-weighted (T2w) MRI sequence was used for this project. The 3D nnU-Net was used as the automatic segmentation framework. The automatic segmentations were compared with the manual segmentations used for clinical practice with Sørensen-Dice coefficient (Dice), 95th Hausdorff distance (95th HD) and mean surface distance (MSD). The dosimetric impact was defined as the difference in D98 (ΔD98) and D90 (ΔD90) between the manual segmentations and the automatic segmentations, evaluated using the clinical dose distribution. The performance of the network was also compared separately depending on FIGO stage and on GTV volume. RESULTS: The network achieved a median Dice of 0.73 (interquartile range (IQR) = 0.50-0.80), median 95th HD of 6.8 mm (IQR = 4.2-12.5 mm) and median MSD of 1.4 mm (IQR = 0.90-2.8 mm). The median ΔD90 and ΔD98 were 0.18 Gy (IQR = -1.38-1.19 Gy) and 0.20 Gy (IQR =-1.10-0.95 Gy) respectively. No significant differences in geometric or dosimetric performance were observed between tumors with different FIGO stages, however significantly improved Dice and dosimetric performance was found for larger tumors. CONCLUSIONS: The nnU-Net framework achieved state-of-the-art performance in the segmentation of the cervical cancer GTV on BT MRI images. Reasonable median performance was achieved geometrically and dosimetrically but with high variability among patients.
引用
- Rodríguez Outeiral R, González PJ, Schaake EE, et al. Deep learning for segmentation of the cervical cancer gross tumor volume on magnetic resonance imaging for brachytherapy. Radiat Oncol. 2023. PMID: 37248490. doi:10.1186/s13014-023-02283-8
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。