Is good old GRAPPA dead?

TL;DR

This paper compares the classical GRAPPA MRI reconstruction method with the deep learning-based XPDNet, analyzing their performance and robustness across different settings to evaluate if traditional methods are still viable.

Contribution

It provides a qualitative analysis of XPDNet's performance relative to GRAPPA, highlighting its generalization capabilities in various MRI reconstruction scenarios.

Findings

XPDNet can generalize well to unseen settings

Deep learning approaches outperform classical methods in certain conditions

Robustness of XPDNet varies across different MRI configurations

Abstract

We perform a qualitative analysis of performance of XPDNet, a state-of-the-art deep learning approach for MRI reconstruction, compared to GRAPPA, a classical approach. We do this in multiple settings, in particular testing the robustness of the XPDNet to unseen settings, and show that the XPDNet can to some degree generalize well.