Exploring Siamese Networks in Self-Supervised Fast MRI Reconstruction
This work addresses the challenge of fast MRI reconstruction without labeled data, offering a practical solution for medical imaging, though it appears incremental as it builds on existing Siamese network concepts.
The paper tackled the problem of reconstructing MR images from undersampled k-space data without fully sampled references by proposing SiamRecon, a self-supervised method using Siamese networks, which achieved state-of-the-art accuracy on brain and knee MRI datasets.
Reconstructing MR images using deep neural networks from undersampled k-space data without using fully sampled training references offers significant value in practice, which is a self-supervised regression problem calling for effective prior knowledge and supervision. The Siamese architectures are motivated by the definition "invariance" and shows promising results in unsupervised visual representative learning. Building homologous transformed images and avoiding trivial solutions are two major challenges in Siamese-based self-supervised model. In this work, we explore Siamese architecture for MRI reconstruction in a self-supervised training fashion called SiamRecon. We show the proposed approach mimics an expectation maximization algorithm. The alternative optimization provide effective supervision signal and avoid collapse. The proposed SiamRecon achieves the state-of-the-art reconstruction accuracy in the field of self-supervised learning on both single-coil brain MRI and multi-coil knee MRI.