Assessment of Data Consistency through Cascades of Independently Recurrent Inference Machines for fast and robust accelerated MRI reconstruction

TL;DR

This paper introduces Cascades of Independently Recurrent Inference Machines (CIRIM), a novel deep learning framework that effectively enforces data consistency in MRI reconstruction, demonstrating superior robustness and generalization in clinical scenarios.

Contribution

The paper proposes CIRIM, a new scheme for assessing data consistency in deep MRI reconstruction networks, comparing explicit and implicit enforcement methods across various architectures.

Findings

CIRIM with implicit data consistency enforcement outperforms other models.

CIRIM maintains lesion contrast in MS patient data with unseen sampling patterns.

CIRIM offers a good balance between image quality and reconstruction speed.

Abstract

Machine Learning methods can learn how to reconstruct Magnetic Resonance Images and thereby accelerate acquisition, which is of paramount importance to the clinical workflow. Physics-informed networks incorporate the forward model of accelerated MRI reconstruction in the learning process. With increasing network complexity, robustness is not ensured when reconstructing data unseen during training. We aim to embed data consistency (DC) in deep networks while balancing the degree of network complexity. While doing so, we will assess whether either explicit or implicit enforcement of DC in varying network architectures is preferred to optimize performance. We propose a scheme called Cascades of Independently Recurrent Inference Machines (CIRIM) to assess DC through unrolled optimization. Herein we assess DC both implicitly by gradient descent and explicitly by a designed term. Extensive comparison of the CIRIM to CS as well as to other methods is performed: the E2EVN, CascadeNet, KIKINet, LPDNet, RIM, IRIM, and UNet. Models were trained and evaluated on T1-weighted and FLAIR contrast brain data, and T2-weighted knee data. Both 1D and 2D undersampling patterns were evaluated. Robustness was tested by reconstructing 7.5x prospectively undersampled 3D FLAIR MRI data of Multiple Sclerosis (MS) patients with white matter lesions. The CIRIM performed best when implicitly enforcing DC, while the E2EVN required an explicit DC formulation. In reconstructing MS patient data, prospectively acquired with a sampling pattern unseen during model training, the CIRIM maintained lesion contrast while efficiently denoising the images. The CIRIM showed highly promising generalization capabilities maintaining a very fair trade-off between reconstructed image quality and fast reconstruction times, which is crucial in the clinical workflow.