Attenuation correction of annihilation photons is essential in PET image reconstruction for providing accurate quantitative activity maps. In the absence of an aligned CT device to obtain attenuation information, we propose the high-resolution residual U-net (HRU-Net) to extract attenuation correction factors (ACF) directly from time-of-flight (TOF) PET emission data. HRU-Net is built upon the U-Net encoding-decoding architecture and it utilizes four blocks of modified residual connections in each stage. In each residual block, concatenation is performed to incorporate input and output feature vectors. In addition, flexible and efficient elements of convolutional neural network (CNN) such as dilated convolutions, pre-activation order of a batch normalization (BN) layer, a rectified linear unit (ReLU) layer and a convolution layer, and residual connections are utilized to extract high resolution features. To illustrate the effectiveness of the proposed method, HRU-Net estimated ACF, attenuation maps and activity maps are compared with maximum likelihood ACF (MLACF) algorithm, U-Net, and HC-Net. An ablation study is conducted using non-TOF and TOF sinograms as inputs of networks. The experimental results show that HRU-Net with TOF projections as inputs leads to normalized root mean square error (NRMSE) of 4.84%1.58%, outperforming MLACF, U-Net and HC-Net with NRMSE of 47.82%13.62%, 6.92%1.94%, and 7.99%2.49%, respectively.

湮没光子的衰减校正在 PET 图像重建中对于提供准确的定量活动图至关重要。在没有对齐的 CT 设备来获取衰减信息的情况下，我们提出了高分辨率残余 U-net (HRU-Net) 以直接从飞行时间 (TOF) PET 发射数据中提取衰减校正因子 (ACF)。HRU-Net 建立在 U-Net 编码-解码架构之上，它在每个阶段使用四个修改后的残差连接块。在每个残差块中，执行连接以合并输入和输出特征向量。此外，卷积神经网络 (CNN) 灵活高效的元素，例如扩张卷积、批量归一化 (BN) 层的预激活顺序、整流线性单元 (ReLU) 层和卷积层，残差连接用于提取高分辨率特征。为了说明所提出方法的有效性，将 HRU-Net 估计的 ACF、衰减图和活动图与最大似然 ACF (MLACF) 算法、U-Net 和 HC-Net 进行了比较。使用非 TOF 和 TOF 正弦图作为网络的输入进行消融研究。实验结果表明，以 TOF 投影作为输入的 HRU-Net 导致归一化均方根误差 (NRMSE) 为 4.84%1.58%，优于 MLACF、U-Net 和 HC-Net，NRMSE 分别为 47.82% 13.62%、6.92% 1.94% 和 7.99% 2.49%。