Unsupervised Medical Image Alignment with Curriculum Learning
This work addresses the challenge of improving accuracy and efficiency in medical image alignment for applications like diagnosis or treatment planning, but it is incremental as it builds on existing state-of-the-art models with curriculum learning adaptations.
The paper tackles the problem of deformable pairwise 3D medical image registration by applying curriculum learning to train convolutional neural networks, starting with easier setups and increasing complexity, and shows that this approach leads to superior results compared to conventional training, with curriculum by input blur achieving the best accuracy versus speed trade-off.
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.