Interdisciplinary Program of Bioengineering, Seoul National University, Seoul, South Korea.
Artificial Intelligence Institute, Seoul National University, Seoul, South Korea.
J Nucl Med. 2024 Oct 1;65(10):1645-1651. doi: 10.2967/jnumed.123.267360.
Quantification of F-FDG PET images is useful for accurate diagnosis and evaluation of various brain diseases, including brain tumors, epilepsy, dementia, and Parkinson disease. However, accurate quantification of F-FDG PET images requires matched 3-dimensional T MRI scans of the same individuals to provide detailed information on brain anatomy. In this paper, we propose a transfer learning approach to adapt a pretrained deep neural network model from amyloid PET to spatially normalize F-FDG PET images without the need for 3-dimensional MRI. The proposed method is based on a deep learning model for automatic spatial normalization of F-FDG brain PET images, which was developed by fine-tuning a pretrained model for amyloid PET using only 103 F-FDG PET and MR images. After training, the algorithm was tested on 65 internal and 78 external test sets. All T MR images with a 1-mm isotropic voxel size were processed with FreeSurfer software to provide cortical segmentation maps used to extract a ground-truth regional SUV ratio using cerebellar gray matter as a reference region. These values were compared with those from spatial normalization-based quantification methods using the proposed method and statistical parametric mapping software. The proposed method showed superior spatial normalization compared with statistical parametric mapping, as evidenced by increased normalized mutual information and better size and shape matching in PET images. Quantitative evaluation revealed a consistently higher SUV ratio correlation and intraclass correlation coefficients for the proposed method across various brain regions in both internal and external datasets. The remarkably good correlation and intraclass correlation coefficient values of the proposed method for the external dataset are noteworthy, considering the dataset's different ethnic distribution and the use of different PET scanners and image reconstruction algorithms. This study successfully applied transfer learning to a deep neural network for F-FDG PET spatial normalization, demonstrating its resource efficiency and improved performance. This highlights the efficacy of transfer learning, which requires a smaller number of datasets than does the original network training, thus increasing the potential for broader use of deep learning-based brain PET spatial normalization techniques for various clinical and research radiotracers.
正电子发射断层扫描（PET）图像的定量分析对于各种脑部疾病的准确诊断和评估非常有用，包括脑肿瘤、癫痫、痴呆和帕金森病。然而，准确的 F-FDG PET 图像定量分析需要对相同个体进行匹配的三维 T 磁共振（MRI）扫描，以提供有关大脑解剖结构的详细信息。在本文中，我们提出了一种转移学习方法，该方法可以将经过预训练的针对淀粉样蛋白 PET 的深度神经网络模型适应于无需三维 MRI 即可进行 F-FDG PET 图像的空间标准化。所提出的方法基于一种用于自动 F-FDG 脑 PET 图像空间标准化的深度学习模型，该模型通过仅使用 103 个 F-FDG PET 和 MR 图像对针对淀粉样蛋白 PET 的预训练模型进行微调而开发。在训练后，算法在 65 个内部和 78 个外部测试集上进行了测试。所有具有 1 毫米各向同性体素大小的 T 磁共振图像均使用 FreeSurfer 软件进行处理，以提供皮质分割图，使用小脑灰质作为参考区域来提取地面真实区域标准化摄取值比。使用所提出的方法和统计参数映射软件将这些值与基于空间标准化的定量方法进行比较。与统计参数映射相比，所提出的方法在空间标准化方面表现出更好的性能，这表现为归一化互信息的增加以及 PET 图像中更好的大小和形状匹配。定量评估表明，在所提出的方法中，对于各种脑区，无论是内部数据集还是外部数据集，标准化摄取值比的相关性和组内相关系数都更高。考虑到外部数据集的不同种族分布以及使用不同的 PET 扫描仪和图像重建算法，所提出的方法对于外部数据集的极好的相关性和组内相关系数值非常值得注意。本研究成功地将转移学习应用于 F-FDG PET 空间标准化的深度神经网络，展示了其资源效率和性能提升。这突显了转移学习的功效，其所需的数据集数量少于原始网络训练，从而增加了基于深度学习的脑 PET 空间标准化技术在各种临床和研究放射性示踪剂中的广泛应用的潜力。
