Noise2Void: unsupervised denoising of PET images.
Noise2Void: unsupervised denoising of PET images.
复制标题
DOI：
10.1088/1361-6560/ac30a0
复制
发表时间：
2021-11-01
影响因子：
3.5
通讯作者：
Dutta J
中科院分区：
文献类型：
--
作者：
Song TA;Yang F;Dutta J
正电子发射断层扫描（PET）图像中的噪声水平升高会降低图像质量和定量准确性，并且是临床解释的混杂因素。深度学习的最新进展带来了一系列新颖的去噪技术，其中一些已成功适用于PET图像重建和后处理。到目前为止，大部分深度学习研究都集中在监督学习方案上，对于图像去噪问题，需要成对的有噪和无噪/低噪图像。这一要求往往限制了这些方法在医学应用中的实用性，因为配对的训练数据集并不总是可用的。此外，为了实现这些方法的最佳情况性能，用于训练和随后的真实世界应用的数据集必须具有一致的图像特性（例如，噪声、分辨率等），这对于临床数据来说是罕见的。为了规避这些挑战，开发无监督技术来满足对配对训练数据的需求至关重要。在本文中，我们采用了Noise 2 Void，一种仅依赖于损坏图像进行模型训练的技术，用于PET图像去噪，并使用PET神经成像数据评估其性能。Noise 2 Void是一种使用盲点网络设计的无监督方法。它只需要一个单一的噪声图像作为其输入，因此，非常适合临床设置。在训练阶段，一个单一的噪声PET图像作为输入和目标。在这里，我们提出了一个基于迁移学习范式的Noise 2 Void的修改版本，该范式涉及组级预训练，然后进行个人微调。此外，我们研究了将解剖图像作为网络的第二输入的影响。我们使用基于BrainWeb数字体模的模拟数据验证了我们的去噪技术。我们表明，具有预训练和/或解剖指导的Noise 2 Void比传统的去噪方案（如高斯滤波，解剖指导的非局部均值滤波，块匹配和4D滤波）具有更高的峰值信噪比。我们使用Noise 2Noise去噪技术作为额外的基准。为了临床验证，我们将此方法应用于人脑成像数据集。临床结果与模拟结果一致，证实了Noise 2 Void作为去噪工具的平移值。
Elevated noise levels in positron emission tomography (PET) images lower image quality and quantitative accuracy and are a confounding factor for clinical interpretation. Recent advances in deep learning have ushered in a wide array of novel denoising techniques, several of which have been successfully adapted for PET image reconstruction and post-processing. The bulk of the deep learning research so far has focused on supervised learning schemes, which, for the image denoising problem, require paired noisy and noiseless/low-noise images. This requirement tends to limit the utility of these methods for medical applications as paired training datasets are not always available. Furthermore, to achieve the best-case performance of these methods, it is essential that the datasets for training and subsequent real-world application have consistent image characteristics (e.g., noise, resolution, etc.), which is rarely the case for clinical data. To circumvent these challenges, it is critical to develop unsupervised techniques that obviate the need for paired training data. In this paper, we have adapted Noise2Void, a technique that relies on corrupt images alone for model training, for PET image denoising and assessed its performance using PET neuroimaging data. Noise2Void is an unsupervised approach that uses a blind-spot network design. It requires only a single noisy image as its input, and, therefore, is well-suited for clinical settings. During the training phase, a single noisy PET image serves as both the input and the target. Here we present a modified version of Noise2Void based on a transfer learning paradigm that involves group-level pretraining followed by individual fine-tuning. Furthermore, we investigate the impact of incorporating an anatomical image as a second input to the network. We validated our denoising technique using simulation data based on the BrainWeb digital phantom. We show that Noise2Void with pretraining and/or anatomical guidance leads to higher peak signal-to-noise ratios than traditional denoising schemes such as Gaussian filtering, anatomically guided non-local means filtering, and block-matching and 4D filtering. We used the Noise2Noise denoising technique as an additional benchmark. For clinical validation, we applied this method to human brain imaging datasets. The clinical findings were consistent with the simulation results confirming the translational value of Noise2Void as a denoising tool.
登录
查看更多内容
影响因子：
19.7
作者：
Chen, Kevin T.;Gong, Enhao;Zaharchuk, Greg
通讯作者：
Zaharchuk, Greg
影响因子：
6.3
作者：
FARDE, L;ERIKSSON, L;HALLDIN, C
通讯作者：
HALLDIN, C
影响因子：
10.6
作者：
Chan, Chung;Fulton, Roger;Meikle, Steven
通讯作者：
Meikle, Steven
影响因子：
12.4
作者：
Dutta J;Ahn S;Li Q
通讯作者：
Li Q
DOI：
10.1109/trpms.2020.2986414
发表时间：
2021-03
影响因子：
4.4
作者：
da Costa-Luis CO;Reader AJ
通讯作者：
Reader AJ