Dynamic-Aware Spatio-temporal Representation Learning for Dynamic MRI Reconstruction

TL;DR

This paper introduces DA-INR, a novel implicit neural representation model for dynamic MRI reconstruction that effectively captures spatio-temporal data, improves reconstruction quality at high undersampling, and reduces optimization time and hyperparameter tuning.

Contribution

The paper presents DA-INR, a new INR-based model that explicitly incorporates temporal redundancy for efficient and high-quality dynamic MRI reconstruction.

Findings

Outperforms existing models at extreme undersampling ratios

Reduces optimization time significantly

Requires minimal hyperparameter tuning

Abstract

Dynamic MRI reconstruction, one of inverse problems, has seen a surge by the use of deep learning techniques. Especially, the practical difficulty of obtaining ground truth data has led to the emergence of unsupervised learning approaches. A recent promising method among them is implicit neural representation (INR), which defines the data as a continuous function that maps coordinate values to the corresponding signal values. This allows for filling in missing information only with incomplete measurements and solving the inverse problem effectively. Nevertheless, previous works incorporating this method have faced drawbacks such as long optimization time and the need for extensive hyperparameter tuning. To address these issues, we propose Dynamic-Aware INR (DA-INR), an INR-based model for dynamic MRI reconstruction that captures the spatial and temporal continuity of dynamic MRI data in the image domain and explicitly incorporates the temporal redundancy of the data into the model structure. As a result, DA-INR outperforms other models in reconstruction quality even at extreme undersampling ratios while significantly reducing optimization time and requiring minimal hyperparameter tuning.