基于迁移学习的生成对抗网络用于静态和动态心脏PET的衰减校正
本研究纳入60例仅静息扫描和14例静息-负荷扫描受试者,基于生成对抗网络从非衰减校正(NAC)PET图像直接生成衰减校正PET图像,用于静息和负荷状态静态或动态[13N]氨水MP PET。结果显示S-DLAC、D-DLAC和D-DLAC-FT方法与临床CTAC性能相当,迁移学习在动态MP PET中展现出潜力。
要点
- S-DLAC与参考静态CTAC的相关性高于静态NAC(斜率0.9423、R2=0.947 vs 斜率0.0992、R2=0.654)。
- D-DLAC-FT在全左心室心肌的MBF误差低于D-DLAC,但无显著差异(静息6.63±5.05% vs 7.00±6.84%,p=0.7593)。
- S-DLAC、D-DLAC和D-DLAC-FT方法与临床CTAC性能相当,迁移学习在动态MP PET中展现出潜力。
核心结论
- 所提出的S-DLAC、D-DLAC和D-DLAC-FT方法与临床CTAC性能相当,迁移学习在动态MP PET中展现出潜力。
研究设计与样本
- 纳入60例仅静息扫描受试者和14例静息-负荷扫描受试者,均接受[13N]氨水心脏PET/CT检查以获取静态和动态帧的3D NAC和CTAC PET图像。
- 开发了基于U-net+ResNet生成器和卷积神经网络判别器的3D pix2pix深度学习衰减校正(DLAC)框架,使用60例仅静息受试者的配对静态或动态NAC和CTAC图像分别训练S-DLAC和D-DLAC。
- 预训练的S-DLAC网络随后用60例仅静息受试者的配对动态NAC和CTAC PET帧进行微调,得到改进的D-DLAC-FT用于动态PET图像。
- 14例静息-负荷受试者作为内部测试数据集,在不同网络模型上分别测试而不参与训练。
主要发现
- S-DLAC(斜率=0.9423,R2=0.947)与参考静态CTAC的相关性高于静态NAC(斜率=0.0992,R2=0.654)。
- D-DLAC-FT在全左心室心肌的MBF误差低于D-DLAC,但无显著差异,60例静息状态受试者为6.63±5.05% vs 7.00±6.84%(p=0.7593)。
- D-DLAC-FT与D-DLAC的MBF误差在14例负荷状态受试者为1.97±2.28% vs 3.21±3.89%(p=0.8595),无显著差异。
关键数值与推荐
- S-DLAC与参考静态CTAC的斜率=0.9423、R2=0.947,静态NAC的斜率=0.0992、R2=0.654。
- D-DLAC-FT与D-DLAC在全左心室心肌的MBF误差:60例静息状态受试者6.63±5.05% vs 7.00±6.84%(p=0.7593)。
- D-DLAC-FT与D-DLAC在全左心室心肌的MBF误差:14例负荷状态受试者1.97±2.28% vs 3.21±3.89%(p=0.8595)。
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: The goal of this work is to demonstrate the feasibility of directly generating attenuation-corrected PET images from non-attenuation-corrected (NAC) PET images for both rest and stress-state static or dynamic [13N]ammonia MP PET based on a generative adversarial network. METHODS: We recruited 60 subjects for rest-only scans and 14 subjects for rest-stress scans, all of whom underwent [13N]ammonia cardiac PET/CT examinations to acquire static and dynamic frames with both 3D NAC and CT-based AC (CTAC) PET images. We developed a 3D pix2pix deep learning AC (DLAC) framework via a U-net + ResNet-based generator and a convolutional neural network-based discriminator. Paired static or dynamic NAC and CTAC PET images from 60 rest-only subjects were used as network inputs and labels for static (S-DLAC) and dynamic (D-DLAC) training, respectively. The pre-trained S-DLAC network was then fine-tuned by paired dynamic NAC and CTAC PET frames of 60 rest-only subjects to derive an improved D-DLAC-FT for dynamic PET images. The 14 rest-stress subjects were used as an internal testing dataset and separately tested on different network models without training. The proposed methods were evaluated using visual quality and quantitative metrics. RESULTS: The proposed S-DLAC, D-DLAC, and D-DLAC-FT methods were consistent with clinical CTAC in terms of various images and quantitative metrics. The S-DLAC (slope = 0.9423, R2 = 0.947) showed a higher correlation with the reference static CTAC as compared to static NAC (slope = 0.0992, R2 = 0.654). D-DLAC-FT yielded lower myocardial blood flow (MBF) errors in the whole left ventricular myocardium than D-DLAC, but with no significant difference, both for the 60 rest-state subjects (6.63 ± 5.05% vs. 7.00 ± 6.84%, p = 0.7593) and the 14 stress-state subjects (1.97 ± 2.28% vs. 3.21 ± 3.89%, p = 0.8595). CONCLUSION: The proposed S-DLAC, D-DLAC, and D-DLAC-FT methods achieve comparable performance with clinical CTAC. Transfer learning shows promising potential for dynamic MP PET.
引用
- Sun H, Wang F, Yang Y, et al. Transfer learning-based attenuation correction for static and dynamic cardiac PET using a generative adversarial network. Eur J Nucl Med Mol Imaging. 2023. PMID: 37474736. doi:10.1007/s00259-023-06343-9
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。