Fast imaging is essential in neonatal brain MRI. Deep learning-based methods can provide high acceleration rates but their performance is instable when limited training data are available. Subspace model-based approach could reduce the dimensionality of imaging and improve reconstruction stability. This work presents a novel method to integrate neonate-specific subspace model and model-driven deep learning, making stable and ultrafast neonatal MR imaging possible. The feasibility and potential of the proposed method have been demonstrated using in vivo data from four medical centers, producing very encouraging results. With further development, the proposed method may provide an effective tool for neonatal imaging.
This abstract and the presentation materials are available to members only; a login is required.