文献速递神经肿瘤

基于MRI的脑膜瘤术前WHO分级预测:多中心深度学习、影像组学、临床及联合模型的开发与比较

MRI-based preoperative WHO grade prediction in meningioma: development and comparison of deep learning, radiomics, clinical, and combined models in multicenter data.
PubMed 文献 · J Neurooncol · 2026-10-14

该多中心回顾性研究纳入455例经病理证实的脑膜瘤,开发并外部验证基于MRI的互补模型及其堆叠组合用于术前WHO分级。多模态集成模型外部AUC最高(0.742),但未显著超过nnU-Net(AUC 0.732);影像组学外部可迁移性差。

要点

  • nnU-Net是外部可迁移性最佳的单预测因子(AUC 0.732)。
  • 多模态集成外部AUC 0.742,未显著优于nnU-Net(ΔAUC +0.011,p=0.737)。
  • 手工影像组学在开发集表现强但外部丢失大部分区分度。

核心结论

  • 多模态集成获得最高外部AUC,但未显著超过nnU-Net,后者是脑膜瘤分级最可迁移的单预测因子。
  • 临床变量未改善区分度,但支持可解释的特征层面归因。
  • 手工影像组学缺乏外部可迁移性。

研究设计与样本

  • 回顾性多中心研究纳入455例经组织病理学证实的脑膜瘤:四中心开发队列(n=346)采用留一中心交叉验证,外部队列(n=109)。
  • 四个基础模型使用对比增强T1加权MRI:肿瘤+瘤周影像组学、基于nnU-Net分割的分类器、冻结DINOv3迁移学习特征及临床变量。
  • 全部15种组合预先设定,使用线性元学习器在开发集折外预测上堆叠;冻结模型未经修改进行外部评估。

主要发现

  • WHO 2-3级肿瘤总体占24.6%(112/455;开发集20.5%;外部集37.6%)。
  • 外部AUC:nnU-Net 0.732,DINOv3 0.703,临床模型0.645,影像组学0.628。
  • 多模态集成(nnU-Net+DINOv3+临床)外部AUC最高(0.742),超过影像组学(ΔAUC +0.114,p=0.008),但未超过nnU-Net(ΔAUC +0.011,p=0.737)。
  • 与临床模型相比,区分度在预设阈值下更高但无统计学意义(ΔAUC +0.098,p=0.050),重分类改善(IDI +0.109,p<0.001)。
  • 含影像组学的组合在开发集表现强,但外部丢失大部分区分度。

临床怎么用

  • 术前无创预测脑膜瘤WHO分级可优先考虑nnU-Net等深度学习模型,其外部可迁移性最佳。
  • 多模态集成虽外部AUC最高,但未显著优于nnU-Net,需权衡复杂度与增益。
  • 手工影像组学模型不宜直接外推至新中心,使用前需外部验证。
  • 临床变量可用于特征层面归因以提升可解释性,但不改善区分度。

局限性

  • 回顾性多中心设计。
  • 外部集WHO 2-3级比例(37.6%)高于开发集(20.5%)。
  • 影像组学外部可迁移性差。
  • 多模态集成较nnU-Net的区分度提升无统计学意义。

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

PURPOSE: The objective of this study was to develop and externally validate complementary magnetic resonance imaging (MRI)-based models and their stacked combinations for preoperative World Health Organization (WHO) grading of intracranial meningioma. METHODS: This retrospective multicenter study included 455 histopathologically confirmed meningiomas: a four-center development cohort (n = 346) with leave-one-center-out cross-validation and an external cohort (n = 109). Four base models used contrast-enhanced T1-weighted MRI: tumor + peritumoral radiomics, an nnU-Net segmentation-based classifier, frozen DINOv3 transfer-learning features, and clinical variables. All 15 combinations were prespecified and stacked using a linear meta-learner trained on development out-of-fold predictions. Frozen models were externally evaluated without modification. RESULTS: WHO 2-3 tumors comprised 24.6% overall (112/455; development 20.5%; external 37.6%). External AUCs were 0.732 for nnU-Net, 0.703 for DINOv3, 0.645 for clinical, 0.628 for radiomics. The multimodal ensemble (nnU-Net+DINOv3+Clinical) achieved the highest external AUC (0.742), exceeding radiomics (ΔAUC +0.114, p = 0.008) but not nnU-Net (ΔAUC +0.011, p = 0.737). Compared with the clinical model, discrimination was higher but nonsignificant at the prespecified threshold (ΔAUC +0.098, p = 0.050), while reclassification improved (IDI +0.109, p < 0.001). Radiomics-containing combinations performed strongly in development but lost most discrimination externally. The multimodal ensemble had an external calibration slope of 1.32 (Brier 0.190) and net benefit above default strategies at thresholds ≥ 0.23. CONCLUSION: The multimodal ensemble achieved the highest external AUC but did not significantly exceed nnU-Net, which was the most transportable single predictor of meningioma grade. Clinical variables did not improve discrimination but enabled interpretable feature-level attribution. Handcrafted radiomics lacked external transportability.

来源:J Neurooncol,查看原文。版权归原作者所有。

引用

  1. Huryol C, Sulaimanov U, Yuce M, et al. MRI-based preoperative WHO grade prediction in meningioma: development and comparison of deep learning, radiomics, clinical, and combined models in multicenter data. J Neurooncol. 2026. PMID: 42875623. doi:10.1007/s11060-026-05827-6

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。