Channel Attention Networks for Robust MR Fingerprinting Matching

TL;DR

This paper introduces a novel neural network with channel attention for improved MR fingerprinting, significantly reducing errors in tissue parameter maps and providing a new channel selection method for better interpretability.

Contribution

It proposes a channel-wise attention neural network architecture and a new attention-based channel selection method for more accurate and explainable MR fingerprinting parameter estimation.

Findings

Reduces T1 error by 8.88%

Reduces T2 error by 75.44%

Analyzes effects of patch size and temporal frames on channel reduction

Abstract

Magnetic Resonance Fingerprinting (MRF) enables simultaneous mapping of multiple tissue parameters such as T1 and T2 relaxation times. The working principle of MRF relies on varying acquisition parameters pseudo-randomly, so that each tissue generates its unique signal evolution during scanning. Even though MRF provides faster scanning, it has disadvantages such as erroneous and slow generation of the corresponding parametric maps, which needs to be improved. Moreover, there is a need for explainable architectures for understanding the guiding signals to generate accurate parametric maps. In this paper, we addressed both of these shortcomings by proposing a novel neural network architecture consisting of a channel-wise attention module and a fully convolutional network. The proposed approach, evaluated over 3 simulated MRF signals, reduces error in the reconstruction of tissue parameters by 8.88% for T1 and 75.44% for T2 with respect to state-of-the-art methods. Another contribution of this study is a new channel selection method: attention-based channel selection. Furthermore, the effect of patch size and temporal frames of MRF signal on channel reduction are analyzed by employing a channel-wise attention.