Real-Time Cardiac Cine MRI with Residual Convolutional Recurrent Neural Network

TL;DR

This paper introduces a novel residual convolutional RNN for real-time cardiac cine MRI reconstruction, enabling faster imaging without ECG gating and outperforming traditional compressed sensing methods.

Contribution

It is the first deep learning approach applied to Cartesian real-time cardiac cine MRI reconstruction, improving image quality and speed.

Findings

Deep learning model outperforms compressed sensing in image quality

First application of deep learning to Cartesian real-time cardiac cine MRI

Radiologists prefer the reconstructed images from the proposed method

Abstract

Real-time cardiac cine MRI does not require ECG gating in the data acquisition and is more useful for patients who can not hold their breaths or have abnormal heart rhythms. However, to achieve fast image acquisition, real-time cine commonly acquires highly undersampled data, which imposes a significant challenge for MRI image reconstruction. We propose a residual convolutional RNN for real-time cardiac cine reconstruction. To the best of our knowledge, this is the first work applying deep learning approach to Cartesian real-time cardiac cine reconstruction. Based on the evaluation from radiologists, our deep learning model shows superior performance than compressed sensing.