文献速递神经核医学肿瘤 PET/CT

基于FDG-PET的稳健残差三维卷积神经网络模型预测β-淀粉样蛋白阳性

A Robust Residual Three-dimensional Convolutional Neural Networks Model for Prediction of Amyloid-β Positivity by Using FDG-PET.
PubMed 文献 · Clin Nucl Med · 2025-06-17

本研究利用187例患者数据开发残差三维卷积神经网络(3DCNN)模型,通过FDG-PET预测β-淀粉样蛋白阳性,并在99例多中心不同种族数据集中评估稳健性。模型在未做站点协调与做站点协调图像上的平均AUC分别为0.815和0.840,F1分数分别为0.770和0.810。作者认为残差3DCNN能学习FDG-PET复杂三维空间模式并稳健预测β-淀粉样蛋白阳性,且显著减少对站点协调预处理的依赖。

要点

  • 模型在未做站点协调图像上平均AUC为0.815,做站点协调后为0.840。
  • 认知正常组AUC在未协调与协调图像上分别为0.801和0.834,痴呆组分别为0.777和0.745。
  • F1分数在认知正常组未协调与协调图像上分别为0.580和0.658,痴呆组分别为0.907和0.931。

核心结论

  • 残差3DCNN可学习FDG-PET图像中复杂的三维空间模式,并稳健预测β-淀粉样蛋白阳性,且显著减少对站点协调预处理的依赖。

研究设计与样本

  • 研究使用187例患者队列开发模型,患者涵盖认知正常至痴呆及其他认知障碍,均接受T1加权MRI、18F-FDG和11C-匹兹堡化合物B(PiB)PET扫描。
  • 采用非穷尽网格搜索配置残差3DCNN模型,并在开发数据集的重复随机划分上训练。
  • 使用99例不同种族、不同站点协调水平图像的多站点数据集评估模型性能及稳健性。

主要发现

  • 模型在未做站点协调与做站点协调图像上的平均AUC分别为0.815和0.840。
  • 认知正常组在未做与做站点协调图像上的AUC分别为0.801和0.834,痴呆组分别为0.777和0.745。
  • F1分数在未做与做站点协调图像上的平均分分别为0.770和0.810。
  • 认知正常组在未做与做站点协调图像上的F1分数分别为0.580和0.658,痴呆组分别为0.907和0.931。

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)。

原文

BACKGROUND: Widely used in oncology PET, 2-deoxy-2- 18 F-FDG PET is more accessible and affordable than amyloid PET, which is a crucial tool to determine amyloid positivity in diagnosis of Alzheimer disease (AD). This study aimed to leverage deep learning with residual 3D convolutional neural networks (3DCNN) to develop a robust model that predicts amyloid-β positivity by using FDG-PET. PATIENTS AND METHODS: In this study, a cohort of 187 patients was used for model development. It consisted of patients ranging from cognitively normal to those with dementia and other cognitive impairments who underwent T1-weighted MRI, 18 F-FDG, and 11 C-Pittsburgh compound B (PiB) PET scans. A residual 3DCNN model was configured using nonexhaustive grid search and trained on repeated random splits of our development data set. We evaluated the performance of our model, and particularly its robustness, using a multisite data set of 99 patients of different ethnicities with images at different site harmonization levels. RESULTS: Our model achieved mean AUC scores of 0.815 and 0.840 on images without and with site harmonization correspondingly. Respectively, it achieved higher AUC scores of 0.801 and 0.834 in the cognitively normal (CN) group compared with 0.777 and 0.745 in the dementia group. As for F1 score, the corresponding mean scores were 0.770 and 0.810 on images without and with site harmonization. In the CN group, it achieved lower F1 scores of 0.580 and 0.658 compared with 0.907 and 0.931 in the dementia group. CONCLUSIONS: We demonstrated that residual 3DCNN can learn complex 3D spatial patterns in FDG-PET images and robustly predict amyloid-β positivity with significantly less reliance on site harmonization preprocessing.

来源:Clin Nucl Med,查看原文。版权归原作者所有。

引用

  1. Ardakani I, Yamada T, Iwano S, et al. A Robust Residual Three-dimensional Convolutional Neural Networks Model for Prediction of Amyloid-β Positivity by Using FDG-PET. Clin Nucl Med. 2025. PMID: 40524364. doi:10.1097/RLU.0000000000005966

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。