儿童弥漫性中线胶质瘤治疗后深度学习分割模型的改进
本研究开发并验证了一种针对儿童弥漫性中线胶质瘤的自动分割工具DMGtracker,基于多机构回顾性数据训练,纳入治疗后影像数据。外部验证显示,DMGtracker在整体肿瘤分割上优于BraTS-PEDs 2024冠军模型,尤其对治疗后扫描的DSC更高、RVD更低。结论认为纳入治疗后扫描训练可显著提升纵向疗效评估中的分割性能。
要点
- DMGtracker用于儿童弥漫性中线胶质瘤整体肿瘤及增强肿瘤的自动分割。
- 训练集中加入治疗后影像可改善整体肿瘤分割的内部DSC。
- 在治疗后扫描中,DMGtracker达到临床可接受整体肿瘤分割(DSC>0.80)的比例为64.0%。
核心结论
- 纳入治疗后扫描训练分割模型可显著改善纵向临床试验影像中的性能,实现更精确的体积追踪和疗效评估。
研究设计与样本
- 多机构回顾性研究,纳入1–30岁弥漫性中线胶质瘤患者,来源包括一家儿科癌症中心、BraTS-PEDs 2024及PNOC007前瞻性试验。
- 训练与内部测试共使用74例患者的153次扫描,其中59次为治疗后扫描。
- 模型基于nnU-Net构建,采用四序列多参数MRI(T1、T1增强、T2、FLAIR)进行训练。
- 训练标注包含140次机构内治疗前后研究的专家分割结果及全部261次BraTS-PEDs 2024治疗前研究。
- 外部验证数据集为88次有标注的PNOC007扫描,来自49例患者,并与BraTS-PEDs 2024冠军模型比较。
诊断标准与关键数值
- 内部测试中,纳入治疗后数据后训练模型的整体肿瘤DSC为0.94(IQR 0.82–0.96),未纳入治疗后数据时为0.93(IQR 0.81–0.96),p<0.001。
- 外部验证中,DMGtracker整体肿瘤分割DSC为0.90(IQR 0.72–0.95),高于BraTS-PEDs 2024冠军模型的0.81(IQR 0.66–0.90)。
- 外部验证中,DMGtracker整体肿瘤分割RVD为9.6%(IQR 3.7%–31.6%),低于对照模型的16.8%(IQR 7.3%–39.2%)。
- 治疗后扫描(n=50)中,DMGtracker整体肿瘤DSC为0.90(IQR 0.73–0.94),对照模型为0.80(IQR 0.58–0.88),p<0.001。
- 治疗后扫描中,DMGtracker整体肿瘤RVD为9.7%(IQR 3.8%–26.2%),对照模型为19.8%(IQR 12.6%–39.8%),p<0.001。
临床应用与阅片要点
- DMGtracker适用于弥漫性中线胶质瘤治疗后的纵向体积追踪,可辅助疗效评估。
- 在治疗后扫描中,DMGtracker达到临床可接受整体肿瘤分割(DSC>0.80)的比例为64.0%,高于对照模型的52.0%。
- 对增强肿瘤的分割同样采用DSC与RVD评估,但原文未给出具体数值,可参考整体肿瘤结果。
局限性
- 研究采用回顾性设计,可能存在选择偏倚。
- 外部验证数据仅来自PNOC007试验,覆盖范围有限。
- 原文未报告增强肿瘤分割的具体DSC与RVD数值。
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)。
英文原文(摘要)· 点击展开
PURPOSE: To develop and validate a pediatric diffuse midline glioma (DMG) auto-segmentation tool optimized for longitudinal treatment response assessment across the disease course.
MATERIALS AND METHODS: In this multi-institutional retrospective study, we included patients aged 1-30 years with DMG from an institutional pediatric cancer center, BraTS-PEDs 2024, and PNOC007, a prospective trial of radiation followed by peptide vaccine plus poly-ICLC, and we trained nnU-Net-based DMGtracker using expert segmentations from 140 institutional pre- and post-treatment studies and all 261 BraTS-PEDs 2024 pre-treatment studies, using four-sequence multiparametric MRI (T1, T1 post-contrast, T2, and FLAIR). We externally validated the model on 88 annotated PNOC007 studies (n = 49 patients) and compared it with the BraTS-PEDs 2024 winning model using median Dice similarity coefficient (DSC) and relative volumetric difference (RVD) for whole-tumor and contrast-enhancing tumor segmentation using the Wilcoxon signed-rank test.
RESULTS: Training and internal testing used 153 scans (59 post-treatment) from 74 patients. Incorporating post-treatment data improved internal whole-tumor DSC for our trained model (0.94 [IQR 0.82-0.96] vs 0.93 [0.81-0.96]; p<0.001). On external validation, DMGtracker outperformed the BraTS-PEDs 2024 winning model for whole-tumor segmentation, with higher DSC (0.90 [0.72-0.95] vs 0.81 [0.66-0.90]) and lower RVD (9.6% [3.7%-31.6%] vs 16.8% [7.3%-39.2%]). This advantage was greatest in post-treatment scans (n = 50 scans, DSC 0.90 [0.73-0.94] vs 0.80 [0.58-0.88]; RVD 9.7% [3.8%-26.2%] vs 19.8% [12.6%-39.8%]; p<0.001 for both). In post-treatment scans, DMGtracker achieved clinically acceptable whole-tumor segmentation (DSC > 0.80) in 64.0% of cases, compared with 52.0% for the BraTS-PEDs winner.
CONCLUSION: Training DMG segmentation models with post-treatment scans substantially improves performance in longitudinal clinical trial imaging, enabling more accurate volumetric tracking and response assessment.
引用
- Zielke, Mussa, Zapaishchykova, et al. Improved Deep Learning Segmentation of Pediatric Diffuse Midline Gliomas After Treatment. AJNR Am J Neuroradiol. 2026. PMID: 42648875. doi:10.3174/ajnr.A9609
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。