Biomechanics-informed Neural Networks for Myocardial Motion Tracking in MRI
Biomechanics-informed Neural Networks for Myocardial Motion Tracking in MRI
复制标题
DOI：
10.1007/978-3-030-59716-0_29
复制
发表时间：
2020-06
期刊：
影响因子：
--
通讯作者：
C. Qin;Shuo Wang;Chen Chen-Chen;Huaqi Qiu;Wenjia Bai;D. Rueckert
中科院分区：
文献类型：
--
作者：
C. Qin;Shuo Wang;Chen Chen-Chen;Huaqi Qiu;Wenjia Bai;D. Rueckert
图像配准是一个不适定的逆问题，通常需要在解空间上正则化。与目前大多数施加显式正则化条件(如光滑性)的方法不同，在本文中，我们提出了一种新的方法，该方法可以隐含地学习生物力学信息的正则化。这种方法可以将特定于应用的先验知识合并到基于深度学习的注册中。具体地说，提出的生物力学信息正则化利用变分自动编码器(VAE)来学习生物力学上可信的变形的流形，并通过重建生物力学模拟来隐含地捕捉它们的潜在属性。然后，学习的VAE正则化可以与任何基于深度学习的注册网络相结合，以将解空间正则化为生物力学上可信的。该方法在两个不同数据集的2D心脏MRI数据堆栈上进行了心肌运动跟踪。结果表明，该方法在运动跟踪精度方面比其他同类方法具有更好的性能，并且具有学习不可压缩和应变等生物力学特性的能力。与常用的L2正则化方案相比，该方法对不可见区域具有更好的泛化能力。
Image registration is an ill-posed inverse problem which often requires regularisation on the solution space. In contrast to most of the current approaches which impose explicit regularisation terms such as smoothness, in this paper we propose a novel method that can implicitly learn biomechanics-informed regularisation. Such an approach can incorporate application-specific prior knowledge into deep learning based registration. Particularly, the proposed biomechanics-informed regularisation leverages a variational autoencoder (VAE) to learn a manifold for biomechanically plausible deformations and to implicitly capture their underlying properties via reconstructing biomechanical simulations. The learnt VAE regulariser then can be coupled with any deep learning based registration network to regularise the solution space to be biomechanically plausible. The proposed method is validated in the context of myocardial motion tracking on 2D stacks of cardiac MRI data from two different datasets. The results show that it can achieve better performance against other competing methods in terms of motion tracking accuracy and has the ability to learn biomechanical properties such as incompressibility and strains. The method has also been shown to have better generalisability to unseen domains compared with commonly used L2 regularisation schemes.