Unsupervised End-to-end Learning for Deformable Medical Image Registration

TL;DR

This paper introduces an unsupervised end-to-end CNN-based registration method for 2D medical images that improves accuracy, speed, and flexibility over traditional techniques, enabling efficient registration of various organs.

Contribution

It adapts traditional registration algorithms into an unsupervised CNN framework, enhancing performance and speed, and allowing training with unlabeled data for diverse organ registration.

Findings

Achieves state-of-the-art results on 2D brain registration.

Speeds up registration by 100x compared to traditional methods.

Improves registration accuracy by approximately 10% with additional data.

Abstract

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.