低计数全身PET/MRI恢复:剂量降低谱系与五种先进人工智能模型的评估
这项多中心研究比较了五种人工智能模型在全身18F-FDG PET/MRI低剂量数据恢复中的表现,覆盖临床标准3 MBq/kg的75%至1%模拟计数水平,并在斯坦福原队列与蒂宾根外部验证队列的476例原始及模拟低计数扫描中评估。结果显示,在剂量降至6.25%时,SwinIR与U-Net可显著改善PET图像诊断质量,但降至临床标准剂量的1%超出当前人工智能技术能力范围。
要点
- 在斯坦福原队列剂量6.25%时,SwinIR的平均SSIM为0.910(95%CI 0.900-0.920),高于EDSR的0.898(0.887-0.910)、EDSR-ViT的0.893(0.881-0.905)、U-Net的0.885(0.873-0.898)和GAN的0.873(0.859-0.887)。
- SwinIR恢复后平均诊断图像质量评分在剂量75%为5(SD 0)、50%为4.50(0.535)、25%为3.75(0.463)、12.5%为3.25(0.463)、6.25%为4(0.926)、1%为2.5(0.534)。
- 与低计数PET图像相比,在较高剂量降低水平(至6.25%)下图像接近或无法诊断时,SwinIR和U-Net均显著改善PET图像诊断质量。
核心结论
- 与低计数PET图像相比,在较高剂量降低水平(至6.25%)下图像接近或无法诊断时,SwinIR和U-Net均显著改善PET图像诊断质量。
- 将放射性示踪剂剂量降至当前临床标准剂量的1%超出当前人工智能技术的能力范围。
研究设计与样本
- 多中心研究,比较五种人工智能模型恢复低计数全身PET/MRI,涵盖U-Net、EDSR、GAN、SwinIR和EDSR-ViT。
- 模型在模拟临床标准3 MBq/kg 18F-FDG剂量的75%、50%、25%、12.5%、6.25%和1%六组计数水平下评估。
- 比较在斯坦福大学原队列与蒂宾根大学跨大陆外部验证队列两个独立队列中进行。
- 共纳入476例原始计数及模拟低计数全身PET/MRI扫描。
主要发现
- 原队列剂量6.25%时平均SSIM:EDSR为0.898(95%CI 0.887-0.910),EDSR-ViT为0.893(0.881-0.905),GAN为0.873(0.859-0.887),U-Net为0.885(0.873-0.898),SwinIR为0.910(0.900-0.920)。
- SwinIR和U-Net的性能还在每个模拟放射性示踪剂剂量水平上进行了审慎评估。
- 斯坦福原队列SwinIR恢复后平均诊断图像质量(DIQ;5点Likert量表)评分:剂量75%为5(SD 0),50%为4.50(0.535),25%为3.75(0.463),12.5%为3.25(0.463),6.25%为4(0.926),1%为2.5(0.534)。
临床怎么用
- 作者认为,在剂量降低至6.25%时,SwinIR和U-Net可显著改善PET图像诊断质量。
- 作者认为,放射性示踪剂剂量降至当前临床标准剂量的1%超出当前人工智能技术能力范围。
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
PURPOSE: To provide a holistic and complete comparison of the five most advanced AI models in the augmentation of low-dose 18F-FDG PET data over the entire dose reduction spectrum. METHODS: In this multicenter study, five AI models were investigated for restoring low-count whole-body PET/MRI, covering convolutional benchmarks - U-Net, enhanced deep super-resolution network (EDSR), generative adversarial network (GAN) - and the most cutting-edge image reconstruction transformer models in computer vision to date - Swin transformer image restoration network (SwinIR) and EDSR-ViT (vision transformer). The models were evaluated against six groups of count levels representing the simulated 75%, 50%, 25%, 12.5%, 6.25%, and 1% (extremely ultra-low-count) of the clinical standard 3 MBq/kg 18F-FDG dose. The comparisons were performed upon two independent cohorts - (1) a primary cohort from Stanford University and (2) a cross-continental external validation cohort from Tübingen University - in order to ensure the findings are generalizable. A total of 476 original count and simulated low-count whole-body PET/MRI scans were incorporated into this analysis. RESULTS: For low-count PET restoration on the primary cohort, the mean structural similarity index (SSIM) scores for dose 6.25% were 0.898 (95% CI, 0.887-0.910) for EDSR, 0.893 (0.881-0.905) for EDSR-ViT, 0.873 (0.859-0.887) for GAN, 0.885 (0.873-0.898) for U-Net, and 0.910 (0.900-0.920) for SwinIR. In continuation, SwinIR and U-Net's performances were also discreetly evaluated at each simulated radiotracer dose levels. Using the primary Stanford cohort, the mean diagnostic image quality (DIQ; 5-point Likert scale) scores of SwinIR restoration were 5 (SD, 0) for dose 75%, 4.50 (0.535) for dose 50%, 3.75 (0.463) for dose 25%, 3.25 (0.463) for dose 12.5%, 4 (0.926) for dose 6.25%, and 2.5 (0.534) for dose 1%. CONCLUSION: Compared to low-count PET images, with near-to or nondiagnostic images at higher dose reduction levels (up to 6.25%), both SwinIR and U-Net significantly improve the diagnostic quality of PET images. A radiotracer dose reduction to 1% of the current clinical standard radiotracer dose is out of scope for current AI techniques.
引用
- Wang YJ, Wang P, Adams LC, et al. Low-count whole-body PET/MRI restoration: an evaluation of dose reduction spectrum and five state-of-the-art artificial intelligence models. Eur J Nucl Med Mol Imaging. 2023. PMID: 36633614. doi:10.1007/s00259-022-06097-w
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。