文献速递胸部MR

基于129Xe与1H MRI的不确定性感知端到端深度学习肺功能量化

Uncertainty-Aware, End-to-End Deep Learning for Functional Lung MRI Quantification Using 129Xe and 1H MRI.
PubMed 文献 · Radiol Cardiothorac Imaging · 2026-01-01

该研究提出一种无需人工干预的端到端深度学习流程,用于从配准的超极化129Xe MRI和结构1H MRI预测通气缺损百分比(VDP),并量化预测不确定性。回顾性研究纳入47名健康者和527例多种肺病患者共574对扫描,采用不确定性感知卷积神经网络与蒙特卡洛dropout。

要点

  • VDP预测中位绝对误差1.01%,无需人工分割,性能可比传统方法
  • 蒙特卡洛dropout提供预测置信区间,20次迭代得到VDP分布及置信分组
  • 临床分类准确率91%,可用于报告VDP分级并提示不确定性

核心结论

  • 临床分类准确率91%(95%CI 68-94,32/35),支持日常阅片辅助使用

研究设计与样本

  • 回顾性研究,时间跨度2015-2024,单中心数据?原文未明确中心数(按原文信息,无法推断)
  • 纳入47名健康者(平均年龄28.3±17.3岁,28名女性)和527例肺病患者(平均年龄44.9±21.9岁,295名女性)共574对129Xe MRI和1H MRI
  • 采用不确定性感知卷积神经网络框架,蒙特卡洛dropout量化不确定性,测试时增强模拟重复性
  • 参考标准为人工校正的分割结果VDP值

关键影像表现/征象

  • 模型输出VDP预测分布,聚类生成置信度分组,可反映预测可靠性
  • VDP预测中位绝对误差1.01%(IQR 0.49-2.47%),误差分布呈低偏差
  • 测试时增强提示模型对配准和噪声具有一定的鲁棒性

诊断标准与关键数值

  • VDP预测中位绝对误差1.01%(IQR 0.49-2.47)
  • 临床分类准确率91%(95%CI 68, 94;32/35),分类标准为VDP严重程度分级(原文未具体分级阈值)
  • 蒙特卡洛dropout 20次迭代生成VDP预测分布用于置信分组

临床应用与阅片要点

  • 报告VDP时可引用自动预测值,并注明置信度信息,辅助临床决策
  • 无需人工轮廓分割,可节省阅片时间,但需配合1H MRI配准质量检查
  • 对VDP异常增高的病例,若置信度低应人工复核分割区域

局限性

  • 原文未明确提及局限性;研究为回顾性,可能受样本选择偏倚影响
  • 不同肺病类型分布可能影响模型泛化,但原文未给出亚组分析
  • 未报告外部验证或前瞻性测试,临床推广需谨慎

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

Purpose To develop an automatic end-to-end deep learning pipeline for predicting the ventilation defect percentage (VDP) from coregistered functional hyperpolarized xenon 129 (129Xe) MRI and structural proton (1H) MRI scans without manual intervention. Materials and Methods In this retrospective study (2015-2024), 129Xe MRI and 1H MRI scans from healthy participants and patients with a range of pulmonary diseases were used to predict VDP and its associated prediction confidence via an uncertainty-aware convolutional neural network framework. Monte Carlo dropout was used to quantify model uncertainty. Model robustness was assessed using test-time augmentation to simulate test-retest repeatability. The proposed approach was evaluated on a stratified testing set via the median absolute error. Results The dataset comprised 574 paired 129Xe MRI and 1H MRI scans from 47 healthy participants (mean ± SD age, 28.3 years ± 17.3; 28 female participants) and 527 patients with a range of pulmonary pathologies (mean ± SD age, 44.9 years ± 21.9; 295 female patients). The proposed framework produced a median absolute error of 1.01% (IQR, 0.49-2.47) VDP compared with manually corrected, segmentation-derived VDPs; no evidence of difference was found (P = .70). Twenty Monte Carlo dropout iterations were completed, producing VDP prediction distributions that were subsequently clustered into confidence groupings. The proposed approach demonstrated clinical classification accuracy of 91% (95% CI: 68, 94; 32 of 35). Conclusion An uncertainty-aware, end-to-end deep learning approach enabled accurate prediction of VDP without manual segmentation, with performance comparable to segmentation-based methods and quantification of prediction uncertainty. Keywords: Functional Imaging, Lung, Multi-Modal, MRI, Uncertainty-Aware Supplemental material is available for this article. © RSNA, 2026.

来源:Radiol Cardiothorac Imaging,查看原文。版权归原作者所有。

引用

  1. Astley, Marshall, Smith, et al. Uncertainty-Aware, End-to-End Deep Learning for Functional Lung MRI Quantification Using 129Xe and 1H MRI. Radiol Cardiothorac Imaging. 2026. PMID: 42312934. doi:10.1148/ryct.250371

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。