Unsupervised Medical Image Alignment with Curriculum Learning

TL;DR

This paper introduces a novel curriculum learning strategy for 3D medical image registration, demonstrating that gradually increasing image clarity during training improves accuracy and efficiency over traditional methods.

Contribution

It is the first to apply curriculum learning to medical image registration, proposing a new input blur approach that enhances model performance and training speed.

Findings

Curriculum learning improves registration accuracy.

Input blur curriculum outperforms other methods in speed-accuracy trade-off.

Proposed method achieves superior results over conventional training.

Abstract

We explore different curriculum learning methods for training convolutional neural networks on the task of deformable pairwise 3D medical image registration. To the best of our knowledge, we are the first to attempt to improve performance by training medical image registration models using curriculum learning, starting from an easy training setup in the first training stages, and gradually increasing the complexity of the setup. On the one hand, we consider two existing curriculum learning approaches, namely curriculum dropout and curriculum by smoothing. On the other hand, we propose a novel and simple strategy to achieve curriculum, namely to use purposely blurred images at the beginning, then gradually transit to sharper images in the later training stages. Our experiments with an underlying state-of-the-art deep learning model show that curriculum learning can lead to superior results compared to conventional training. Additionally, we show that curriculum by input blur has the best accuracy versus speed trade-off among the compared curriculum learning approaches.