Multi-Weight Respecification of Scan-specific Learning for Parallel Imaging

TL;DR

This paper introduces MW-RAKI, a multi-weight neural network approach for parallel MRI reconstruction that improves noise resilience and high acceleration performance over existing methods.

Contribution

The paper proposes MW-RAKI, a novel multi-weighting strategy integrated into RAKI, enhancing noise robustness and high acceleration imaging in parallel MRI reconstruction.

Findings

MW-RAKI outperforms traditional RAKI at high acceleration rates.

Multi-weighting reduces noise influence and improves data constraints.

Experimental results show superior reconstruction quality with MW-RAKI.

Abstract

Parallel imaging is widely used in magnetic resonance imaging as an acceleration technology. Traditional linear reconstruction methods in parallel imaging often suffer from noise amplification. Recently, a non-linear robust artificial-neural-network for k-space interpolation (RAKI) exhibits superior noise resilience over other linear methods. However, RAKI performs poorly at high acceleration rates, and needs a large amount of autocalibration signals as the training samples. In order to tackle these issues, we propose a multi-weight method that implements multiple weighting matrices on the undersampled data, named as MW-RAKI. Enforcing multiple weighted matrices on the measurements can effectively reduce the influence of noise and increase the data constraints. Furthermore, we incorporate the strategy of multiple weighting matrixes into a residual version of RAKI, and form MW-rRAKI.Experimental compari-sons with the alternative methods demonstrated noticeably better reconstruction performances, particularly at high acceleration rates.