Sparse-view and limited-angle CT reconstruction with untrained networks and deep image prior
Sparse-view and limited-angle CT reconstruction with untrained networks and deep image prior
复制标题
DOI：
10.1016/j.cmpb.2022.107167
复制
发表时间：
2022-10
影响因子：
6.1
通讯作者：
Ziyu Shu;A. Entezari
中科院分区：
文献类型：
--
作者：
Ziyu Shu;A. Entezari
背景与目的：基于神经网络的图像重建方法越来越流行。然而，有限的训练数据和缺乏理论保证的普遍性引起了关注，特别是在生物医学成像应用中。众所周知，这些挑战导致重建过程不稳定，这对生物医学图像重建构成了重大问题。在本文中，我们提出了一个新的框架，该框架使用未经训练的生成器网络来应对这一挑战，利用深度网络的结构来基于深度图像先验（DIP）技术的正则化解决方案。方法：为了获得较高的重建精度，我们提出了一个框架，在重建过程中对潜在向量和生成器网络的权重进行优化。为了提高框架的稳定性和收敛性能，我们还提出了相应的重构策略。此外，我们提出将其法算子实现为平行光束几何下的卷积核，而不是每次迭代都计算正演投影，从而大大加快了计算速度。结果：我们的实验表明，在稀疏视图、有限角度和低剂量条件下，所提出的框架比其他最先进的传统、预训练和未训练方法有显著改进。结论：应用于平行束x射线成像，我们的框架在重建过程的速度、准确性和稳定性方面具有优势。我们还表明，所提出的框架与生物医学图像重建文献中常用的所有可微分正则化兼容。我们的框架也可以用作后处理技术，进一步改进任何其他重建方法生成的重建。此外，所提出的框架不需要训练数据，可以按需调整以适应不同的条件（例如噪声水平、几何形状和成像对象）。
Background and objective:Neural network based image reconstruction methods are becoming increasingly popular. However, limited training data and the lack of theoretical guarantees for generalizability raised concerns, especially in biomedical imaging applications. These challenges are known to lead to an unstable reconstruction process that poses significant problems in biomedical image reconstruction. In this paper, we present a new framework that uses untrained generator networks to tackle this challenge, leveraging the structure of deep networks for regularizing solutions based on a technique known as Deep Image Prior (DIP).Methods:To achieve a high reconstruction accuracy, we propose a framework optimizing both the latent vector and the weights of a generator network during the reconstruction process. We also propose the corresponding reconstruction strategies to improve the stability and convergent performance of the proposed framework. Furthermore, instead of calculating forward projection in each iteration, we propose implementing its normal operator as a convolutional kernel under parallel beam geometry, thus greatly accelerating the calculation.Results:Our experiments show that the proposed framework has significant improvements over other state-of-the-art conventional, pre-trained, and untrained methods under sparse-view, limited-angle, and low-dose conditions.Conclusions:Applying to parallel beam X-ray imaging, our framework shows advantages in speed, accuracy, and stability of the reconstruction process. We also show that the proposed framework is compatible with all differentiable regularizations that are commonly used in biomedical image reconstruction literature. Our framework can also be used as a post-processing technique to further improve the reconstruction generated by any other reconstruction methods. Furthermore, the proposed framework requires no training data and can be adjusted on-demand to adapt to different conditions (e.g. noise level, geometry, and imaged object).