Multi-scale MRI reconstruction via dilated ensemble networks

TL;DR

This paper presents a multi-scale MRI reconstruction network using dilated convolutions and complex-valued features, achieving superior performance and efficiency over existing methods by preserving resolution and phase information.

Contribution

Introduces a novel multi-scale MRI reconstruction network with dilated convolutions and complex features, improving detail recovery and efficiency compared to prior architectures.

Findings

Real-valued model outperforms existing architectures and is three times more efficient.

Complex-valued network yields better qualitative results with phase information.

Multi-scale approach effectively captures both global and local features.

Abstract

As aliasing artefacts are highly structural and non-local, many MRI reconstruction networks use pooling to enlarge filter coverage and incorporate global context. However, this inadvertently impedes fine detail recovery as downsampling creates a resolution bottleneck. Moreover, real and imaginary features are commonly split into separate channels, discarding phase information particularly important to high frequency textures. In this work, we introduce an efficient multi-scale reconstruction network using dilated convolutions to preserve resolution and experiment with a complex-valued version using complex convolutions. Inspired by parallel dilated filters, multiple receptive fields are processed simultaneously with branches that see both large structural artefacts and fine local features. We also adopt dense residual connections for feature aggregation to efficiently increase scale and the deep cascade global architecture to reduce overfitting. The real-valued version of this model outperformed common reconstruction architectures as well as a state-of-the-art multi-scale network whilst being three times more efficient. The complex-valued network yielded better qualitative results when more phase information was present.