Unsupervised End-to-end Learning for Deformable Medical Image Registration
This work addresses faster and more accurate registration of medical images like brain and liver for healthcare applications, but it is incremental as it adapts traditional methods to a neural network framework.
The paper tackles the problem of deformable medical image registration for 2D CT/MRI images by proposing an unsupervised end-to-end convolutional neural network, achieving state-of-the-art results on brain registration and comparable results on liver registration, with a 100x speed improvement over traditional methods and a 10% performance boost from additional unlabeled data.
We propose a registration algorithm for 2D CT/MRI medical images with a new unsupervised end-to-end strategy using convolutional neural networks. The contributions of our algorithm are threefold: (1) We transplant traditional image registration algorithms to an end-to-end convolutional neural network framework, while maintaining the unsupervised nature of image registration problems. The image-to-image integrated framework can simultaneously learn both image features and transformation matrix for registration. (2) Training with additional data without any label can further improve the registration performance by approximately 10 %. (3) The registration speed is 100x faster than traditional methods. The proposed network is easy to implement and can be trained efficiently. Experiments demonstrate that our system achieves state-of-the-art results on 2D brain registration and achieves comparable results on 2D liver registration. It can be extended to register other organs beyond liver and brain such as kidney, lung, and heart.