文献速递神经MR

自动弥散分析无创预测WHO 2-3级胶质瘤异柠檬酸脱氢酶基因型

Automated Diffusion Analysis for Noninvasive Prediction of Isocitrate Dehydrogenase Genotype in WHO Grade 2-3 Gliomas.
PubMed 文献 · AJNR Am J Neuroradiol · 2025-10-01

本研究基于UCLH医院247例胶质瘤数据,训练nnUNet深度学习算法仅用T2加权序列分割肿瘤并提取ADC直方图特征,用于预测IDH基因型;nnUNet分割中位Dice为0.85(BraTS数据)和0.83(UCLH数据),最佳指标归一化ADCmean的AUC为0.82(95%CI 0.78-0.88),与人工分割AUC 0.84(95%CI 0.77-0.89)统计学等效(P<.01)。

要点

  • nnUNet仅用T2序列自动分割胶质瘤并提取ADC,其IDH预测性能与人工分割统计学等效(P<.01)。
  • 最佳指标归一化ADCmean鉴别IDH突变型与野生型的AUC为0.82(95%CI 0.78-0.88)。
  • nnUNet发现1处人工观察者遗漏的胶质瘤浸润区,0.8%病例漏判肿瘤成分,6%病例出现肿瘤远处脑组织过度分割。

核心结论

  • 仅用T2加权序列训练的nnUNet算法所提取的ADC读数用于IDH基因分型,其性能与人工观察者统计学等效,该流程可能支持在疾病早期基于ADC快速识别胶质母细胞瘤,即使输入数据有限。

研究设计与样本

  • 研究基于UCLH医院数据集247例胶质瘤,人工在T2加权MRI上分割并配准至ADC图提取直方图;nnUNet深度学习算法仅用BraTS 2021数据(n=500,80%训练、5%验证、15%测试)训练分割,再应用于UCLH数据;采用单变量逻辑回归检验ADC指标预测IDH状态,配对双侧t检验评估统计学等效性。

关键影像表现/征象

  • nnUNet分割在BraTS数据中位Dice为0.85,在UCLH数据中位Dice为0.83。
  • nnUNet发现1处人工观察者遗漏的胶质瘤浸润区。
  • 0.8%的胶质瘤中nnUNet漏判肿瘤成分。
  • 6%的病例出现远离肿瘤的脑组织过度分割(如颞极)。

诊断标准与关键数值

  • 归一化ADCmean鉴别IDH突变型与IDH野生型胶质瘤的AUC为0.82(95%CI 0.78-0.88),人工分割AUC为0.84(95%CI 0.77-0.89)。
  • 所有ADC指标中,人工与nnUNet提取的ADC值统计学等效(P<.01)。

临床应用与阅片要点

  • 本流程基于T2加权序列自动分割并提取ADC,可辅助IDH基因型无创预测,减少人工分割耗时与操作者依赖性。
  • 阅片时需注意nnUNet可能漏判肿瘤成分(0.8%)及远离肿瘤处过度分割(6%,如颞极),人工复核有助于发现遗漏浸润区。

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)。

英文原文(摘要)· 点击展开

BACKGROUND AND PURPOSE: Glioma molecular characterization is essential for risk stratification and treatment planning. Noninvasive imaging biomarkers such as ADC values have shown potential for predicting glioma genotypes. However, manual segmentation of gliomas is time-consuming and operator-dependent. To address this limitation, we aimed to establish a single-sequence-derived automatic ADC extraction pipeline by using T2-weighted imaging to support glioma isocitrate dehydrogenase (IDH) genotyping.

MATERIALS AND METHODS: Glioma volumes from a hospital data set (University College London Hospitals [UCLH]; n = 247) were manually segmented on T2-weighted MRI scans by using ITK-Snap Toolbox and coregistered to ADC map sequences by using the FMRIB Linear Image Registration Tool in FSL, followed by ADC histogram extraction (Python). Separately, a nnUNet deep learning algorithm was trained to segment glioma volumes by using T2-weighted sequences only from Brain Tumor Segmentation Challenge (BraTS) 2021 data (n = 500, 80% training, 5% validation, and 15% test split). nnUNet was then applied to the UCLH data for segmentation and ADC readouts. Univariable logistic regression was used to test the performance manual and nnUNet derived ADC metrics for IDH status prediction. Statistical equivalence was tested (paired 2-sided t test).

RESULTS: nnUNet segmentation achieved a median Dice of 0.85 on BraTS data, and 0.83 on UCLH data. For the best performing metric (normalized ADCmean ) the area under the receiver operating characteristic curve (AUC) for differentiating IDH-mutant from IDH-wild-type gliomas was 0.82 (95% CI, 0.78-0.88), compared with the manual segmentation AUC 0.84 (95% CI, 0.77-0.89). For all ADC metrics, manually and nnUNet-extracted ADC were statistically equivalent (P < .01). nnUNet identified 1 area of glioma infiltration missed by human observers. In 0.8% gliomas, nnUNet missed glioma components. In 6% of cases, oversegmentation of brain remote from the tumor occurred (eg, temporal poles).

CONCLUSIONS: The T2-weighted trained nnUNet algorithm achieved ADC readouts for IDH genotyping with a performance statistically equivalent to human observers. This approach could support rapid ADC-based identification of glioblastoma at an early disease stage, even with limited input data. Artificial intelligence level of evidence: 5A.

来源:AJNR Am J Neuroradiol,查看原文。版权归原作者所有。

引用

  1. Wu J, Thust SC, Wastling SJ, et al. Automated Diffusion Analysis for Noninvasive Prediction of Isocitrate Dehydrogenase Genotype in WHO Grade 2-3 Gliomas. AJNR Am J Neuroradiol. 2025. PMID: 40494626. doi:10.3174/ajnr.A8776

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。