Leube Julian, Gustafsson Johan, Lassmann Michael, Salas-Ramirez Maikol, Tran-Gia Johannes
Department of Nuclear Medicine, University of Würzburg, Oberdürrbacher Str. 6, 97080, Würzburg, Germany.
Medical Radiation Physics, Lund, Lund University, Skåne University Hospital, Lund, 221 85, Lund, Sweden.
EJNMMI Phys. 2022 Jul 19;9(1):47. doi: 10.1186/s40658-022-00476-w.
In recent years, a lot of effort has been put in the enhancement of medical imaging using artificial intelligence. However, limited patient data in combination with the unavailability of a ground truth often pose a challenge to a systematic validation of such methodologies. The goal of this work was to investigate a recently proposed method for an artificial intelligence-based generation of synthetic SPECT projections, for acceleration of the image acquisition process based on a large dataset of realistic SPECT simulations.
A database of 10,000 SPECT projection datasets of heterogeneous activity distributions of randomly placed random shapes was simulated for a clinical SPECT/CT system using the SIMIND Monte Carlo program. Synthetic projections at fixed angular increments from a set of input projections at evenly distributed angles were generated by different u-shaped convolutional neural networks (u-nets). These u-nets differed in noise realization used for the training data, number of input projections, projection angle increment, and number of training/validation datasets. Synthetic projections were generated for 500 test projection datasets for each u-net, and a quantitative analysis was performed using statistical hypothesis tests based on structural similarity index measure and normalized root-mean-squared error. Additional simulations with varying detector orbits were performed on a subset of the dataset to study the effect of the detector orbit on the performance of the methodology. For verification of the results, the u-nets were applied to Jaszczak and NEMA physical phantom data obtained on a clinical SPECT/CT system.
No statistically significant differences were observed between u-nets trained with different noise realizations. In contrast, a statistically significant deterioration was found for training with a small subset (400 datasets) of the 10,000 simulated projection datasets in comparison with using a large subset (9500 datasets) for training. A good agreement between synthetic (i.e., u-net generated) and simulated projections before adding noise demonstrates a denoising effect. Finally, the physical phantom measurements show that our findings also apply for projections measured on a clinical SPECT/CT system.
Our study shows the large potential of u-nets for accelerating SPECT/CT imaging. In addition, our analysis numerically reveals a denoising effect when generating synthetic projections with a u-net. Clinically interesting, the methodology has proven robust against camera orbit deviations in a clinically realistic range. Lastly, we found that a small number of training samples (e.g., ~ 400 datasets) may not be sufficient for reliable generalization of the u-net.
近年来，人们在利用人工智能增强医学成像方面付出了诸多努力。然而，有限的患者数据以及缺乏真实情况往往给此类方法的系统验证带来挑战。这项工作的目标是研究一种最近提出的基于人工智能生成合成单光子发射计算机断层扫描（SPECT）投影的方法，以基于大量真实SPECT模拟数据集加速图像采集过程。
使用SIMIND蒙特卡罗程序为临床SPECT/CT系统模拟了一个包含10000个具有随机放置的随机形状的异质活性分布的SPECT投影数据集的数据库。由不同的U形卷积神经网络（U-net）从一组均匀分布角度的输入投影以固定角度增量生成合成投影。这些U-net在用于训练数据的噪声实现、输入投影数量、投影角度增量以及训练/验证数据集数量方面存在差异。为每个U-net为500个测试投影数据集生成合成投影，并使用基于结构相似性指数测量和归一化均方根误差的统计假设检验进行定量分析。在数据集的一个子集上进行了具有不同探测器轨道的额外模拟，以研究探测器轨道对该方法性能的影响。为验证结果，将U-net应用于在临床SPECT/CT系统上获得的Jaszczak和NEMA物理体模数据。
在使用不同噪声实现训练的U-net之间未观察到统计学上的显著差异。相比之下，与使用10000个模拟投影数据集中的大子集（9500个数据集）进行训练相比，使用小子集（400个数据集）进行训练发现有统计学上的显著恶化。在添加噪声之前，合成（即U-net生成）投影与模拟投影之间的良好一致性表明了去噪效果。最后，物理体模测量表明我们的发现也适用于在临床SPECT/CT系统上测量的投影。
我们的研究表明U-net在加速SPECT/CT成像方面具有巨大潜力。此外，我们的分析从数值上揭示了使用U-net生成合成投影时的去噪效果。在临床上有趣的是，该方法在临床实际范围内已被证明对相机轨道偏差具有鲁棒性。最后，我们发现少量的训练样本（例如，约400个数据集）可能不足以使U-net进行可靠的泛化。
