Radiation Physics, Skåne University Hospital, Malmö, Sweden
Eigenvision AB, Malmö, Sweden.
J Nucl Med. 2020 Feb;61(2):298-303. doi: 10.2967/jnumed.119.226613. Epub 2019 Jul 19.
Scintillation camera images contain a large amount of Poisson noise. We have investigated whether noise can be removed in whole-body bone scans using convolutional neural networks (CNNs) trained with sets of noisy and noiseless images obtained by Monte Carlo simulation. : Three CNNs were generated using 3 different sets of training images: simulated bone scan images, images of a cylindric phantom with hot and cold spots, and a mix of the first two. Each training set consisted of 40,000 noiseless and noisy image pairs. The CNNs were evaluated with simulated images of a cylindric phantom and simulated bone scan images. The mean squared error between filtered and true images was used as difference metric, and the coefficient of variation was used to estimate noise reduction. The CNNs were compared with gaussian and median filters. A clinical evaluation was performed in which the ability to detect metastases for CNN- and gaussian-filtered bone scans with half the number of counts was compared with standard bone scans. : The best CNN reduced the coefficient of variation by, on average, 92%, and the best standard filter reduced the coefficient of variation by 88%. The best CNN gave a mean squared error that was on average 68% and 20% better than the best standard filters, for the cylindric and bone scan images, respectively. The best CNNs for the cylindric phantom and bone scans were the dedicated CNNs. No significant differences in the ability to detect metastases were found between standard, CNN-, and gaussian-filtered bone scans. Noise can be removed efficiently regardless of noise level with little or no resolution loss. The CNN filter enables reducing the scanning time by half and still obtaining good accuracy for bone metastasis assessment.
闪烁相机图像包含大量的泊松噪声。我们研究了是否可以使用通过蒙特卡罗模拟获得的带有噪声和无噪声图像的集合来训练卷积神经网络（CNN），以去除全身骨扫描中的噪声。使用 3 种不同的训练图像集生成了 3 个 CNN：模拟骨扫描图像、带有热点和冷点的圆柱形幻影图像以及前两者的混合图像。每个训练集由 40000 对无噪声和噪声图像组成。使用圆柱形幻影的模拟图像和模拟骨扫描图像评估了 CNN。将滤波后图像和真实图像之间的均方误差用作差异度量，并使用变异系数估计降噪。将 CNN 与高斯和中值滤波器进行了比较。进行了临床评估，比较了使用计数减少一半的 CNN 和高斯滤波的骨扫描检测转移的能力与标准骨扫描。最好的 CNN 平均降低了 92%的变异系数，而最好的标准滤波器平均降低了 88%的变异系数。对于圆柱形和骨扫描图像，最好的 CNN 的均方误差平均分别比最好的标准滤波器好 68%和 20%。圆柱形幻影和骨扫描的最佳 CNN 是专用 CNN。在检测转移的能力方面，标准、CNN 和高斯滤波的骨扫描之间没有发现显著差异。无论噪声水平如何，都可以有效地去除噪声，而分辨率损失很小或没有。CNN 滤波器可以将扫描时间减半，并仍然获得良好的骨转移评估准确性。
