Unsupervised learning of probabilistic diffeomorphic registration for images and surfaces
Unsupervised learning of probabilistic diffeomorphic registration for images and surfaces
复制标题
DOI：
10.1016/j.media.2019.07.006
复制
发表时间：
2019-10-01
影响因子：
10.9
通讯作者：
Sabuncu, Mert R.
中科院分区：
文献类型：
--
作者：
Dalca, Adrian V.;Balakrishnan, Guha;Sabuncu, Mert R.
经典的可变形配准技术取得了令人印象深刻的结果，并提供了严格的理论处理，但计算密集，因为它们解决了每个图像对的优化问题。最近，基于学习的方法通过学习空间变形函数来促进快速配准。然而，这些方法使用受限的变形模型，需要监督标签，或者不保证拓扑保持的配准。此外，基于学习的配准工具还没有从概率框架，可以提供不确定性estimations.In本文中，我们建立了经典和基于学习的方法之间的连接。我们提出了一个概率生成模型，并推导出一个基于无监督学习的推理算法，该算法使用了经典配准方法的见解，并利用了卷积神经网络（CNN）的最新发展。我们证明了我们的方法在3D大脑注册任务的图像和解剖表面，并提供了广泛的经验分析的算法。我们的原则性方法导致了最先进的精度和非常快的运行时间，同时提供了同构保证。我们的实现可以在http://voxelmorph.csail.mit.edu上获得。(C)2019 Elsevier B.V.版权所有。
Classical deformable registration techniques achieve impressive results and offer a rigorous theoretical treatment, but are computationally intensive since they solve an optimization problem for each image pair. Recently, learning-based methods have facilitated fast registration by learning spatial deformation functions. However, these approaches use restricted deformation models, require supervised labels, or do not guarantee a diffeomorphic (topology-preserving) registration. Furthermore, learning-based registration tools have not been derived from a probabilistic framework that can offer uncertainty estimates.In this paper, we build a connection between classical and learning-based methods. We present a probabilistic generative model and derive an unsupervised learning-based inference algorithm that uses insights from classical registration methods and makes use of recent developments in convolutional neural networks (CNNs). We demonstrate our method on a 3D brain registration task for both images and anatomical surfaces, and provide extensive empirical analyses of the algorithm. Our principled approach results in state of the art accuracy and very fast runtimes, while providing diffeomorphic guarantees. Our implementation is available online at http://voxelmorph.csail.mit.edu. (C) 2019 Elsevier B.V. All rights reserved.