Department of Radiological, Technology, Faculty of Medical Technology, Teikyo University, 2-11-1 Kaga, Itabashi-ku, Tokyo 173-8605, Japan.
Department of Radiology, Saiseikai Yokohamashi Tobu Hospital, 3-6-1 Shimosueyoshi, Tsurumi-ku, Yokohama, Kanagawa 230-0012, Japan.
Phys Med. 2022 Aug;100:18-25. doi: 10.1016/j.ejmp.2022.06.006. Epub 2022 Jun 15.
Deep-layer learning processing may improve contrast imaging with greater precision in low-count acquisition. However, no data on noise reduction using super-resolution processing for deep-layer learning have been reported in nuclear medicine imaging.
This study was designed to evaluate the adaptability of deep denoising super-resolution convolutional neural networks (DDSRCNN) in nuclear medicine by comparing them with denoising convolutional natural networks (DnCNN), Gaussian processing, and nonlinear diffusion (NLD) processing.
In this study, 156 patients were included. Data were collected using a matrix size of 256 × 256 with a pixel size of 2.46 mm at 0.898 folds, 15% energy window at the center of the photopeak energy (140 keV), and total count of 1000 kilocounts (kct). Following the training and validation of two learning models, we created 100 images for each 20-test datum. The peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) between each image and the reference image were calculated.
DDSRCNN showed the highest PSNR values for all total counts. Regarding SSIM, DDSRCNN had significantly higher values than the original and Gaussian. In DnCNN, false accumulation was observed as the total counts increased. Regarding PSNR and SSIM transition, the model using 100-500-kct training data was significantly higher than that using 100-kct training data.
Edge-preserving noise reduction processing was possible, and adaptability to low-count acquisition was demonstrated using DDSRCNN. Using training data with different noise levels, DDSRCNN could learn the noise components with high accuracy and contrast improvement.
深度学习处理可能会提高低计数采集的对比成像精度。然而，在核医学成像中，尚未有关于使用超分辨率处理进行深度降噪的报道。
本研究旨在通过与去卷积神经网络（DnCNN）、高斯处理和非线性扩散（NLD）处理进行比较，评估深度降噪超分辨率卷积神经网络（DDSRCNN）在核医学中的适应性。
本研究纳入了 156 名患者。使用矩阵大小为 256×256、像素大小为 2.46mm、0.898 倍放大率、15%能量窗在光电峰能量（140keV）中心、总计数为 1000 千计数（kct）的设备采集数据。在训练和验证两个学习模型后，我们为每个 20 个测试数据创建了 100 张图像。计算每张图像与参考图像之间的峰值信噪比（PSNR）和结构相似性（SSIM）。
在所有总计数中，DDSRCNN 显示出最高的 PSNR 值。关于 SSIM，DDSRCNN 的值明显高于原始值和高斯值。在 DnCNN 中，随着总计数的增加，出现了错误的积累。关于 PSNR 和 SSIM 的转换，使用 100-500-kct 训练数据的模型明显高于使用 100-kct 训练数据的模型。
可以进行边缘保持降噪处理，DDSRCNN 可实现对低计数采集的适应性。使用具有不同噪声水平的训练数据，DDSRCNN 可以高精度地学习噪声分量并提高对比度。
