Nuclear Engineering Department, Shiraz University, Shiraz, Iran.
Division of Nuclear Medicine and Molecular Imaging, Department of Medical Imaging, Geneva University Hospital, CH-1211, Geneva 4, Switzerland.
J Digit Imaging. 2023 Aug;36(4):1588-1596. doi: 10.1007/s10278-023-00815-y. Epub 2023 Mar 29.
The existing deep learning-based denoising methods predicting standard-dose PET images (S-PET) from the low-dose versions (L-PET) solely rely on a single-dose level of PET images as the input of deep learning network. In this work, we exploited the prior knowledge in the form of multiple low-dose levels of PET images to estimate the S-PET images. To this end, a high-resolution ResNet architecture was utilized to predict S-PET images from 6 to 4% L-PET images. For the 6% L-PET imaging, two models were developed; the first and second models were trained using a single input of 6% L-PET and three inputs of 6%, 4%, and 2% L-PET as input to predict S-PET images, respectively. Similarly, for 4% L-PET imaging, a model was trained using a single input of 4% low-dose data, and a three-channel model was developed getting 4%, 3%, and 2% L-PET images. The performance of the four models was evaluated using structural similarity index (SSI), peak signal-to-noise ratio (PSNR), and root mean square error (RMSE) within the entire head regions and malignant lesions. The 4% multi-input model led to improved SSI and PSNR and a significant decrease in RMSE by 22.22% and 25.42% within the entire head region and malignant lesions, respectively. Furthermore, the 4% multi-input network remarkably decreased the lesions' SUV bias and SUV bias by 64.58% and 37.12% comparing to single-input network. In addition, the 6% multi-input network decreased the RMSE within the entire head region, within the lesions, lesions' SUV bias, and SUV bias by 37.5%, 39.58%, 86.99%, and 45.60%, respectively. This study demonstrated the significant benefits of using prior knowledge in the form of multiple L-PET images to predict S-PET images.
现有的基于深度学习的去噪方法仅依赖于单次剂量水平的 PET 图像作为深度学习网络的输入，来预测标准剂量 PET 图像（S-PET）。在这项工作中，我们利用了以多种低剂量 PET 图像形式呈现的先验知识来估计 S-PET 图像。为此，我们利用了高分辨率 ResNet 架构，从 6%到 4%的低剂量 PET 图像预测 S-PET 图像。对于 6%的低剂量 PET 成像，我们开发了两种模型；第一种和第二种模型分别使用单一的 6%低剂量 PET 输入和 6%、4%和 2%低剂量 PET 的三个输入来训练，以预测 S-PET 图像。同样，对于 4%的低剂量 PET 成像，我们使用单一的 4%低剂量数据输入训练了一个模型，并开发了一个三通道模型，用于获取 4%、3%和 2%的低剂量 PET 图像。我们使用结构相似性指数（SSI）、峰值信噪比（PSNR）和均方根误差（RMSE）来评估这四个模型在整个头部区域和恶性病变中的性能。4%的多输入模型在整个头部区域和恶性病变中导致 SSI 和 PSNR 显著提高，而 RMSE 显著降低 22.22%和 25.42%。此外，4%的多输入网络与单输入网络相比，显著降低了病变的 SUV 偏差和 SUV 偏差，分别为 64.58%和 37.12%。此外，6%的多输入网络降低了整个头部区域、病变内、病变的 SUV 偏差和 SUV 偏差的 RMSE，分别为 37.5%、39.58%、86.99%和 45.60%。这项研究表明，利用多种低剂量 PET 图像形式的先验知识来预测 S-PET 图像具有显著的优势。
