文献速递神经MR

基于深度学习的常规MRI提升脑转移瘤、胶质母细胞瘤与原发性中枢神经系统淋巴瘤的鉴别诊断

Improving MRI-based differentiation of brain metastasis, glioblastoma, and primary CNS lymphoma using deep learning.
PubMed 文献 · Neuroradiology · 2026-03-06

这项多中心回顾性研究纳入1298例经组织病理学确诊的脑转移瘤、胶质母细胞瘤或原发性中枢神经系统淋巴瘤患者,基于轴位T2WI和增强T1WI构建2.5D ResNet50深度学习模型并与临床模型比较。深度学习模型内部测试AUC为0.893、外部为0.863,在胶质母细胞瘤和淋巴瘤分类上优于临床模型(均p<0.001),放射科医生借助深度学习后平均准确率由69.8%升至81.4%(内部)和由73.8%升至87.8%(外部,均p<0.05)。

要点

  • 深度学习模型基于轴位T2WI和增强T1WI,可在术前鉴别脑转移瘤、胶质母细胞瘤与原发性中枢神经系统淋巴瘤。
  • 外部测试集AUC为0.863,胶质母细胞瘤与淋巴瘤的一对多AUC显著高于临床模型(均p<0.001)。
  • 放射科医生借助深度学习后平均诊断准确率内部由69.8%升至81.4%、外部由73.8%升至87.8%(均p<0.05)。

核心结论

  • 基于常规MRI的深度学习模型能够准确鉴别脑转移瘤、胶质母细胞瘤和原发性中枢神经系统淋巴瘤,整合到影像判读中可带来额外诊断获益。

研究设计与样本

  • 多中心回顾性研究,纳入1298例经组织病理学确诊的患者(脑转移瘤426例、胶质母细胞瘤456例、原发性中枢神经系统淋巴瘤416例),基于轴位T2WI和增强T1WI构建2.5D ResNet50深度学习模型,并以多变量logistic回归构建纳入人口学与语义影像特征的临床模型,在独立内部和外部测试集评估。

诊断标准与关键数值

  • 深度学习模型鉴别三类肿瘤的AUC内部测试集为0.893(95%CI 0.773–1.000),外部测试集为0.863(95%CI 0.862–0.864)。
  • 胶质母细胞瘤和原发性中枢神经系统淋巴瘤的一对多AUC在内部和外部数据集中均显著高于临床模型(均p<0.001)。
  • 脑转移瘤的一对多AUC在外部数据集中与临床模型无显著差异(p=0.97)。
  • 放射科医生平均准确率借助深度学习后内部由69.8%升至81.4%、外部由73.8%升至87.8%(均p<0.05)。

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

OBJECTIVE: To develop and validate a deep learning (DL) model based on conventional MRI for preoperative differentiation of brain metastasis (BM), glioblastoma (GBM), and primary central nervous system lymphoma (PCNSL), and to evaluate its diagnostic utility as a decision-support tool for radiologists. METHODS: This multicenter retrospective study included 1,298 patients with histopathologically confirmed BM (n = 426), GBM (n = 456), or PCNSL (n = 416). A 2.5D ResNet50-based DL model was developed using axial T2-weighted and contrast-enhanced T1-weighted MRI. A clinical model (CM) incorporating demographic and semantic imaging features was constructed using multivariable logistic regression. Model performance was evaluated in independent internal and external test sets using the area under the receiver operating characteristic curve (AUC) and accuracy with 95% confidence intervals (CIs). Pairwise AUC comparisons were performed using the DeLong test. A reader study involving three radiologists with different experience levels compared diagnostic accuracy before and after DL assistance. RESULTS: The DL model achieved AUCs of 0.893 (95% CI: 0.773–1.000) internally and 0.863 (95% CI: 0.862–0.864) externally. One-versus-rest AUCs were significantly higher than those of the CM for GBM and PCNSL in both datasets (all p < 0.001), whereas no significant difference was observed for BM externally (p = 0.97). With DL assistance, mean radiologist accuracy increased from 69.8% to 81.4% internally and from 73.8% to 87.8% externally (all p < 0.05). CONCLUSION: The proposed DL model based on conventional MRI enables accurate differentiation of BM, GBM, and PCNSL and provides additive diagnostic benefit when integrated into radiological interpretation.

来源:Neuroradiology,查看原文。版权归原作者所有。

引用

  1. Du X, Jin Y, Chen S, et al. Improving MRI-based differentiation of brain metastasis, glioblastoma, and primary CNS lymphoma using deep learning. Neuroradiology. 2026. PMID: 41790161. doi:10.1007/s00234-026-03960-7

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。