Ultrasound Elasticity Imaging Using Physics-Based Models and Learning-Based Plug-and-Play Priors
Ultrasound Elasticity Imaging Using Physics-Based Models and Learning-Based Plug-and-Play Priors
复制标题
DOI：
10.1109/icassp39728.2021.9413652
复制
发表时间：
2021-03
期刊：
影响因子：
--
通讯作者：
N. Mohammadi;M. Doyley;M. Çetin
中科院分区：
文献类型：
--
作者：
N. Mohammadi;M. Doyley;M. Çetin
现有的基于物理模型的超声弹性重建成像方法利用固定的变分正则化器，其可能不适合于感兴趣的应用，或者可能不捕获关于底层组织的复杂空间先验信息。另一方面，基于端到端学习的方法仅依赖于训练数据，而没有利用成像系统的管理物理定律。将基于学习的先验知识与超声弹性成像的物理正向模型相结合，提出了一种联合重建框架，该框架保证了学习驱动的重建与底层物理相一致。为了将弹性逆问题作为正则化优化问题来解决，我们提出了一种即插即用（Pestival）重建方法，其中弹性图像估计过程的每次迭代都涉及单独的更新，包括数据保真度和基于学习的正则化。在这种方法中，数据保真度项是使用一个统计线性代数模型的准静态平衡方程揭示了所观察到的位移场的关系，未观察到的弹性模量。正则化器包括基于卷积神经网络（CNN）的去噪器，其捕获底层组织的学习的先验结构。初步的仿真结果表明，所提出的方法的鲁棒性和有效性有限的训练数据集和噪声位移测量。
Existing physical model-based imaging methods for ultrasound elasticity reconstruction utilize fixed variational regularizers that may not be appropriate for the application of interest or may not capture complex spatial prior information about the underlying tissues. On the other hand, end-to-end learning-based methods count solely on the training data, not taking advantage of the governing physical laws of the imaging system. Integrating learning-based priors with physical forward models for ultrasound elasticity imaging, we present a joint reconstruction framework which guarantees that learning driven reconstructions are consistent with the underlying physics. For solving the elasticity inverse problem as a regularized optimization problem, we propose a plug-and-play (PnP) reconstruction approach in which each iteration of the elasticity image estimation process involves separate updates incorporating data fidelity and learning-based regularization. In this methodology, the data fidelity term is developed using a statistical linear algebraic model of quasi-static equilibrium equation revealing the relationship of the observed displacement fields to the unobserved elastic modulus. The regularizer comprises a convolutional neural network (CNN) based denoiser that captures the learned prior structure of the underlying tissues. Preliminary simulation results demonstrate the robustness and effectiveness of the proposed approach with limited training datasets and noisy displacement measurements.