Pan Boyang, Qi Na, Meng Qingyuan, Wang Jiachen, Peng Siyue, Qi Chengxiao, Gong Nan-Jie, Zhao Jun
RadioDynamic Healthcare, Shanghai, China.
Department of Nuclear Medicine, Shanghai East Hospital, Tongji University School of Medicine, No. 150 Jimo Road, Pudong New District, Shanghai, China.
EJNMMI Phys. 2022 Jun 13;9(1):43. doi: 10.1186/s40658-022-00472-0.
To generate high-quality bone scan SPECT images from only 1/7 scan time SPECT images using deep learning-based enhancement method.
Normal-dose (925-1110 MBq) clinical technetium 99 m-methyl diphosphonate (99mTc-MDP) SPECT/CT images and corresponding SPECT/CT images with 1/7 scan time from 20 adult patients with bone disease and a phantom were collected to develop a lesion-attention weighted U-Net (Qin et al. in Pattern Recognit 106:107404, 2020), which produces high-quality SPECT images from fast SPECT/CT images. The quality of synthesized SPECT images from different deep learning models was compared using PSNR and SSIM. Clinic evaluation on 5-point Likert scale (5 = excellent) was performed by two experienced nuclear physicians. Average score and Wilcoxon test were constructed to assess the image quality of 1/7 SPECT, DL-enhanced SPECT and the standard SPECT. SUVmax, SUVmean, SSIM and PSNR from each detectable sphere filled with imaging agent were measured and compared for different images.
U-Net-based model reached the best PSNR (40.8) and SSIM (0.788) performance compared with other advanced deep learning methods. The clinic evaluation showed the quality of the synthesized SPECT images is much higher than that of fast SPECT images (P < 0.05). Compared to the standard SPECT images, enhanced images exhibited the same general image quality (P > 0.999), similar detail of 99mTc-MDP (P = 0.125) and the same diagnostic confidence (P = 0.1875). 4, 5 and 6 spheres could be distinguished on 1/7 SPECT, DL-enhanced SPECT and the standard SPECT, respectively. The DL-enhanced phantom image outperformed 1/7 SPECT in SUVmax, SUVmean, SSIM and PSNR in quantitative assessment.
Our proposed method can yield significant image quality improvement in the noise level, details of anatomical structure and SUV accuracy, which enabled applications of ultra fast SPECT bone imaging in real clinic settings.
使用基于深度学习的增强方法，仅从1/7扫描时间的SPECT图像生成高质量骨扫描SPECT图像。
收集20例成年骨病患者和一个体模的正常剂量（925 - 1110 MBq）临床锝99m - 亚甲基二膦酸盐（99mTc - MDP）SPECT/CT图像以及扫描时间为其1/7的相应SPECT/CT图像，以开发一种病变注意力加权U - Net（Qin等人，《模式识别》，106:107404，2020），该模型可从快速SPECT/CT图像生成高质量SPECT图像。使用PSNR和SSIM比较不同深度学习模型合成的SPECT图像质量。由两名经验丰富的核医学医师进行5分制李克特量表（5 = 优秀）的临床评估。构建平均分和威尔科克森检验以评估1/7 SPECT、深度学习增强SPECT和标准SPECT的图像质量。测量并比较不同图像中每个充满显像剂的可检测球体的SUVmax、SUVmean、SSIM和PSNR。
与其他先进的深度学习方法相比，基于U - Net的模型达到了最佳的PSNR（40.8）和SSIM（0.788）性能。临床评估表明，合成的SPECT图像质量远高于快速SPECT图像（P < 0.05）。与标准SPECT图像相比，增强图像表现出相同的总体图像质量（P > 0.999）、相似的99mTc - MDP细节（P = 0.125）和相同的诊断置信度（P = 0.1875）。在1/7 SPECT、深度学习增强SPECT和标准SPECT上分别可区分4、5和6个球体。在定量评估中，深度学习增强的体模图像在SUVmax、SUVmean、SSIM和PSNR方面优于1/7 SPECT。
我们提出的方法可在噪声水平、解剖结构细节和SUV准确性方面显著提高图像质量，这使得超快速SPECT骨成像能够在实际临床环境中应用。
