文献速递神经MR

SAGE 栖息地分析鉴别三类颅内肿瘤的多中心研究

Differentiation of Pathologically Distinct Intracranial Tumors Using SAGE-Based Habitat Analysis: A Multicenter Study.
PubMed 文献 · Acad Radiol · 2026-08-03

本研究回顾性纳入 90 例内部开发队列(3.0T 与 5.0T)和 28 例外部队列(3.0T),评估多参数 SAGE 成像的栖息地分析能否整合空间异质性,用于脑膜瘤、脑转移瘤与胶质瘤的三分类鉴别。K=3 被选为最简洁可重复的栖息地表征,Mean+Habitat K3+RBF-SVM 内部宏观 AUC 为 0.700±0.086,外部宏观 AUC 为 0.629。

要点

  • SAGE 血管构筑图 K=3 栖息地分割可提取空间异质性特征。
  • 外部验证中 Mean Only+线性 SVM 宏观 AUC 高达 0.849。
  • SHAP 分析显示全局均值与栖息地特征均有贡献。

核心结论

  • SAGE 栖息地分析提供了空间分辨特征,用于三类颅内肿瘤(脑膜瘤、脑转移瘤、胶质瘤)的术前鉴别。

研究设计与样本

  • 回顾性多中心研究,内部开发队列 90 例(3.0T 及 5.0T 扫描),外部队列 28 例(3.0T 扫描)。
  • SAGE 血管构筑图采用 K-means 聚类分割,候选 K 值从 2 到 6。
  • 特征类别包括全局均值参数、全瘤放射组学和栖息地衍生放射组学,组合为七个特征集。
  • 经 ComBat 协调后,10 个分类器采用 30 次重复分层蒙特卡洛交叉验证评估,最高内部验证宏观 AUC 模型进行外部验证。

关键影像表现/征象

  • SAGE 参数图可生成血管构筑图,提供肿瘤内部空间异质性信息。

诊断标准与关键数值

  • K=3 被选为最简洁且可重复的栖息地表征。
  • Mean + Habitat K3 + 径向基函数支持向量机内部验证平均宏观 AUC 为 0.700±0.086,外部宏观 AUC 为 0.629。
  • 事后比较中,Mean Only + 线性支持向量机外部宏观 AUC 为 0.849。

临床应用与阅片要点

  • 多参数 SAGE 栖息地分析可作为颅内肿瘤术前无创三分类的辅助影像工具。
  • 特征重要性分析提示阅片时应同时关注全局均值参数和空间异质性栖息地特征。

局限性

  • 回顾性研究设计存在偏倚风险。
  • 外部验证中集成栖息地特征模型的宏观 AUC 低于仅用全局均值特征的线性 SVM 模型。

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

RATIONALE AND OBJECTIVES: Preoperative differentiation of meningiomas, brain metastases, and gliomas remains challenging because of overlapping magnetic resonance imaging features and intratumoral heterogeneity. We evaluated whether habitat analysis of multiparametric spin-and-gradient-echo (SAGE) imaging could incorporate spatial heterogeneity into three-class tumor differentiation.

MATERIALS AND METHODS: This retrospective multicenter study included an internal development cohort of 90 patients scanned at 3.0 T and 5.0 T and an external cohort of 28 patients scanned at 3.0 T. SAGE-derived vascular architecture maps were partitioned by K-means clustering across candidate K values from 2 to 6. Three feature categories were extracted from SAGE parameter maps: global mean parameters, whole-tumor radiomics, and habitat-derived radiomics; their combinations yielded seven feature sets. After ComBat harmonization, 10 classifiers were evaluated using 30-repetition stratified Monte Carlo cross-validation. The model with the highest mean internal validation macro-AUC was evaluated externally. Shapley additive explanations (SHAP) analysis of a prespecified secondary tree-based model examined feature contributions.

RESULTS: K=3 was selected as the most parsimonious and reproducible habitat representation. Mean + Habitat K3 + radial-basis-function support vector machine achieved the highest mean internal validation macro-AUC (0.700±0.086). Its external macro-AUC was 0.629. In a post hoc comparison, Mean Only + Linear support vector machine achieved an external macro-AUC of 0.849. SHAP analysis indicated contributions from global mean and habitat-derived features.

CONCLUSION: SAGE-based habitat analysis provided spatially resolved features for three-class intracranial tumor differentiation.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Li, Chen, Li, et al. Differentiation of Pathologically Distinct Intracranial Tumors Using SAGE-Based Habitat Analysis: A Multicenter Study. Acad Radiol. 2026. PMID: 42547382. doi:10.1016/j.acra.2026.07.027

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。