文献速递神经肿瘤

深度学习模型自动提取PET RANO标准:外部验证应用于[18F]FDOPA PET

Automatic extraction of PET RANO criteria with an externally validated deep learning model: Application to [18F]FDOPA PET imaging.
PubMed 文献 · Neuro Oncol · 2026-01-01

本研究回顾性纳入3个欧洲中心635例[18F]FDOPA PET扫描,训练3D U-Net分割肿瘤并自动提取PET RANO 1.0参数。模型在外部验证和测试集Dice分别为0.885和0.851,MTV、TBRmax、TBRmean与专家定量一致性高(ICC>0.93),可测量病灶识别率超97%。

要点

  • 3D U-Net在[18F]FDOPA PET上自动分割胶质瘤,外部测试集Dice为0.851。
  • 自动提取的MTV、TBRmax、TBRmean与专家定量高度一致(ICC>0.93)。
  • 可测量病灶识别正确率超过97%,支持氨基酸PET定量评估自动化。

核心结论

  • 该[18F]FDOPA PET深度学习模型具有稳健的多中心性能,可实现全自动、可重复的定量分析。

研究设计与样本

  • 回顾性纳入3个欧洲中心共635例静态[18F]FDOPA PET扫描,用于胶质瘤诊断、复发评估或治疗监测。
  • 训练集为Nancy医院530例扫描,外部验证集为Pitié-Salpêtrière医院66例,测试集为Turin医院39例。
  • 参考标准分割遵循国际指南,训练3D U-Net分割肿瘤和健康脑体积。

主要发现

  • 肿瘤分割Dice在训练集为0.925(0.841; 0.970),验证集为0.885(0.829; 0.925),测试集为0.851(0.733; 0.911)。
  • 病灶水平与专家定量一致性高:MTV偏差2.293 [-4.734; 9.321] mL,TBRmax偏差0.056 [-0.189; 0.301],TBRmean偏差-0.139 [-0.424; 0.146]。
  • 组内相关系数均大于0.93,可测量病灶正确识别率超过97%。

关键数值与推荐

  • 以全肿瘤体积计算Dice系数评估分割性能。
  • 在病灶水平评估PET RANO 1.0参数(TBRmean、TBRmax、MTV)的定量一致性。

临床怎么用

  • 该模型可自动提取[18F]FDOPA PET的PET RANO标准参数,辅助胶质瘤治疗监测中的定量评估。
  • 模型已公开(https://github.com/IADI-Nancy/FDOPA-PET-GliomaSeg),支持氨基酸PET在神经肿瘤中的更广泛应用。

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

BACKGROUND: Automatic segmentation of gliomas on amino acid PET is essential for quantitative tumor assessment, a pillar in monitoring gliomas under treatment. This study aimed to develop a deep learning model for the automated extraction of PET RANO criteria from [18F]FDOPA PET, with external validation. METHODS: A total of 635 static [18F]FDOPA PET scans from 3 European centers were retrospectively included for glioma diagnosis, recurrence assessment, or treatment monitoring. The training cohort comprised 530 scans from Nancy Hospital, with external validation and test sets from Pitié-Salpêtrière Hospital (n = 66) and Turin Hospital (n = 39). Ground-truth segmentations followed international guidelines. A 3D U-Net was trained to segment tumor and healthy brain volumes. Performance was evaluated using the Dice coefficient using the whole tumor volume. Quantitative agreement for PET RANO criteria 1.0 parameters, tumor-to-background ratios (TBRmean, TBRmax) and metabolic tumor volume (MTV), was assessed at the lesion level. RESULTS: Tumor segmentation achieved Dice of 0.925 (0.841; 0.970) in training, 0.885 (0.829; 0.925) in validation, and 0.851 (0.733; 0.911) in the test set. At lesion level, agreement with expert quantification was high, with low bias and strong reliability for MTV (2.293 [-4.734; 9.321] mL), TBRmax (0.056 [-0.189; 0.301]), and TBRmean (-0.139 [-0.424; 0.146]) and intraclass correlation coefficients superior to 0.93. Measurable lesions were correctly identified in more than 97% of cases. CONCLUSION: Our [18F]FDOPA PET deep learning model (available at https://github.com/IADI-Nancy/FDOPA-PET-GliomaSeg) demonstrates robust multicenter performance and enables fully automated, reproducible quantification, supporting broader clinical adoption of amino acid PET in neuro-oncology.

来源:Neuro Oncol,查看原文。版权归原作者所有。

引用

  1. Zaragori T, Rozenblum L, Rovera G, et al. Automatic extraction of PET RANO criteria with an externally validated deep learning model: Application to [18F]FDOPA PET imaging. Neuro Oncol. 2026. PMID: 42089130. doi:10.1093/neuonc/noag095

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。