指南与共识泌尿生殖技术与安全PETCT

深度学习自动测量[68Ga]PSMA-11或[18F]DCFPyL、[18F]FDG及[177Lu]Lu-PSMA-617显像中SUV与分子肿瘤体积的全局阈值区域共识网络

Deep Learning for Automated Measures of SUV and Molecular Tumor Volume in [68Ga]PSMA-11 or [18F]DCFPyL, [18F]FDG, and [177Lu]Lu-PSMA-617 Imaging with Global Threshold Regional Consensus Network.
指南 / 共识 · J Nucl Med · 2025-11-03

本研究基于机构队列的676例[68Ga]PSMA-11或[18F]DCFPyL PET、390例[18F]FDG PET及477例LuPSMA SPECT图像,开发了基于nnU-Net的全局阈值区域共识网络用于自动勾画疾病范围,并在56例外院PET/CT分期病例上外部测试。模型在PSMA PET、FDG PET和LuPSMA SPECT的体素Dice相似系数分别为0.94、0.84和0.97,外部测试Dice准确率在PSMA和FDG PET分别为0.95和0.84,定量生物标志物与人工参考标准相关性高。结论认为自动深度学习方法可高精度勾画疾病范围和示踪剂亲和力,并提供预训练模型和脚本供开放使用。

要点

  • 全局阈值区域共识网络在PSMA PET的Dice为0.94,FDG PET为0.84,LuPSMA SPECT为0.97。
  • 外部测试PSMA PET Dice准确率0.95,FDG PET为0.84。
  • 定量生物标志物与人工参考标准Pearson相关系数在0.88至0.99之间。

核心结论

  • 自动深度学习方法可高精度勾画疾病范围和示踪剂亲和力。
  • 结合阈值后处理,工具有望接近人工工作流程的输出。
  • 预训练模型和适应机构数据的脚本已提供开放使用。

研究设计与样本

  • 基于机构队列的PET/CT和定量SPECT/CT图像,共676例[68Ga]PSMA-11或[18F]DCFPyL PET、390例[18F]FDG PET和477例LuPSMA SPECT用于开发自动工作流程。
  • 在56例外院转诊PET/CT分期病例上进行外部测试,参考标准为人工标注的总疾病负荷。
  • 分割框架为基于nnU-Net的全局阈值区域共识网络,并加入处理优化以改善边界定义和整体标签准确性。

诊断标准与关键数值

  • 模型勾画疾病范围的体素Dice相似系数:[68Ga]PSMA-11或[18F]DCFPyL PET为0.94,[18F]FDG PET为0.84,LuPSMA SPECT为0.97。
  • 外部测试Dice准确率:PSMA PET为0.95,FDG PET为0.84。
  • 与人工参考标准比较的Pearson相关系数:疾病体积在[68Ga]PSMA-11或[18F]DCFPyL、[18F]FDG和LuPSMA分别为0.98、0.94、0.99;SUVmean分别为0.98、0.88、0.99;SUVmax分别为0.96、0.91、0.99;体积强度乘积分别为0.97、0.96、0.99。

临床应用与阅片要点

  • 原文未据此给出推荐。

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

Metastatic castration-resistant prostate cancer has a high rate of mortality with a limited number of effective treatments after hormone therapy. Radiopharmaceutical therapy with [177Lu]Lu-prostate-specific membrane antigen-617 (LuPSMA) is one treatment option; however, response varies and is partly predicted by PSMA expression and metabolic activity, assessed on [68Ga]PSMA-11 or [18F]DCFPyL and [18F]FDG PET, respectively. Automated methods to measure these on PET imaging have previously yielded modest accuracy. Refining computational workflows and standardizing approaches may improve patient selection and prognostication for LuPSMA therapy. Methods: PET/CT and quantitative SPECT/CT images from an institutional cohort of patients staged for LuPSMA therapy were annotated for total disease burden. In total, 676 [68Ga]PSMA-11 or [18F]DCFPyL PET, 390 [18F]FDG PET, and 477 LuPSMA SPECT images were used for development of automated workflow and tested on 56 cases with externally referred PET/CT staging. A segmentation framework, the Global Threshold Regional Consensus Network, was developed based on nnU-Net, with processing refinements to improve boundary definition and overall label accuracy. Results: Using the model to contour disease extent, the mean volumetric Dice similarity coefficient for [68Ga]PSMA-11 or [18F]DCFPyL PET was 0.94, for [18F]FDG PET was 0.84, and for LuPSMA SPECT was 0.97. On external test cases, Dice accuracy was 0.95 and 0.84 on PSMA and FDG PET, respectively. The refined models yielded consistent improvements compared with nnU-Net, with an increase of 3%-5% in Dice accuracy and 10%-17% in surface agreement. Quantitative biomarkers were compared with a human-defined ground truth using the Pearson coefficient, with scores for [68Ga]PSMA-11 or [18F]DCFPyL, [18F]FDG, and LuPSMA, respectively, of 0.98, 0.94, and 0.99 for disease volume; 0.98, 0.88, and 0.99 for SUVmean; 0.96, 0.91, and 0.99 for SUVmax; and 0.97, 0.96, and 0.99 for volume intensity product. Conclusion: Delineation of disease extent and tracer avidity can be performed with a high degree of accuracy using automated deep learning methods. By incorporating threshold-based postprocessing, the tools can closely match the output of manual workflows. Pretrained models and scripts to adapt to institutional data are provided for open use.

来源:J Nucl Med,查看原文。版权归原作者所有。

引用

  1. Jackson P, Buteau JP, McIntosh L, et al. Deep Learning for Automated Measures of SUV and Molecular Tumor Volume in [68Ga]PSMA-11 or [18F]DCFPyL, [18F]FDG, and [177Lu]Lu-PSMA-617 Imaging with Global Threshold Regional Consensus Network. J Nucl Med. 2025. PMID: 40967759. doi:10.2967/jnumed.125.270077

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。