Department of Nuclear Medicine, Päijät-Häme Central Hospital, Lahti, Finland.
HERMES Medical Solutions, Stockholm, Sweden.
Biomed Phys Eng Express. 2023 Sep 13;9(6). doi: 10.1088/2057-1976/acf66c.
. The quality of myocardial perfusion SPECT (MPS) images is often hampered by low count statistics. Poor image quality might hinder reporting the studies and in the worst case lead to erroneous diagnosis. Deep learning (DL)-based methods can be used to improve the quality of the low count studies. DL can be applied in several different methods, which might affect the outcome. The aim of this study was to investigate the differences between post reconstruction- and reconstruction-based denoising methods.. A UNET-type network was trained using ordered subsets expectation maximization (OSEM) reconstructed MPS studies acquired with half, quarter and eighth of full-activity. The trained network was applied as a post reconstruction denoiser (OSEM+DL) and it was incorporated into a regularized reconstruction algorithm as a deep learning penalty (DLP). OSEM+DL and DLP were compared against each other and against OSEM images without DL denoising in terms of noise level, myocardium-ventricle contrast and defect detection performance with signal-to-noise ratio of a non-prewhitening matched filter (NPWMF-SNR) applied to artificial perfusion defects inserted into defect-free clinical MPS scans. Comparisons were made using half-, quarter- and eighth-activity data.. OSEM+DL provided lower noise level at all activities than other methods. DLP's noise level was also always lower than matching activity OSEM's. In addition, OSEM+DL and DLP outperformed OSEM in defect detection performance, but contrary to noise level ranking DLP had higher NPWMF-SNR overall than OSEM+DL. The myocardium-ventricle contrast was highest with DLP and lowest with OSEM+DL. Both OSEM+DL and DLP offered better image quality than OSEM, but visually perfusion defects were deeper in OSEM images at low activities.. Both post reconstruction- and reconstruction-based DL denoising methods have great potential for MPS. The preference between these methods is a trade-off between smoother images and better defect detection performance.
. 心肌灌注单光子发射计算机断层成像术（SPECT）图像的质量通常受到低计数统计数据的影响。较差的图像质量可能会阻碍报告研究，在最坏的情况下导致错误的诊断。基于深度学习（DL）的方法可用于改善低计数研究的质量。DL 可以以几种不同的方式应用，这可能会影响结果。本研究旨在探讨重建后和基于重建的去噪方法之间的差异。. 使用有序子集期望最大化（OSEM）重建的 MPS 研究，以半剂量、四分之一剂量和八分之一剂量采集，训练了一种 UNET 型网络。训练好的网络作为后重建去噪器（OSEM+DL）应用，并作为正则化重建算法中的深度学习惩罚（DLP）纳入其中。OSEM+DL 和 DLP 与 OSEM 图像进行了比较，这些 OSEM 图像没有进行基于 DL 的去噪，比较的指标是噪声水平、心肌与心室的对比度以及使用非预白化匹配滤波器（NPWMF-SNR）的人工灌注缺陷的检测性能，NPWMF-SNR 应用于无缺陷的临床 MPS 扫描中插入的缺陷扫描。比较是在半剂量、四分之一剂量和八分之一剂量数据下进行的。. OSEM+DL 在所有活动中提供的噪声水平均低于其他方法。DLP 的噪声水平也始终低于匹配活动 OSEM 的噪声水平。此外，OSEM+DL 和 DLP 在缺陷检测性能方面优于 OSEM，但与噪声水平排名相反，DLP 的 NPWMF-SNR 总体上高于 OSEM+DL。DLP 的心肌与心室对比度最高，OSEM+DL 的对比度最低。OSEM+DL 和 DLP 均提供了比 OSEM 更好的图像质量，但在低活动下，OSEM 图像中的灌注缺陷看起来更深。. 基于重建后和基于重建的 DL 去噪方法都具有很大的 MPS 潜力。这些方法之间的偏好是平滑图像和更好的缺陷检测性能之间的权衡。
