Spatial Attention-based Implicit Neural Representation for Arbitrary Reduction of MRI Slice Spacing

TL;DR

This paper introduces a novel spatial attention-based implicit neural network that can reconstruct MRI images at arbitrary slice spacings, addressing the limitations of fixed-scale super-resolution methods in clinical scenarios.

Contribution

The proposed SA-INR model enables flexible MRI slice spacing reduction using implicit neural representations with local-aware spatial attention and efficiency improvements.

Findings

Outperforms existing methods on public and clinical datasets.

Capable of arbitrary inter-slice spacing reconstruction.

Improves computational efficiency with gradient-guided gating.

Abstract

Magnetic resonance (MR) images collected in 2D clinical protocols typically have large inter-slice spacing, resulting in high in-plane resolution and reduced through-plane resolution. Super-resolution technique can enhance the through-plane resolution of MR images to facilitate downstream visualization and computer-aided diagnosis. However, most existing works train the super-resolution network at a fixed scaling factor, which is not friendly to clinical scenes of varying inter-slice spacing in MR scanning. Inspired by the recent progress in implicit neural representation, we propose a Spatial Attention-based Implicit Neural Representation (SA-INR) network for arbitrary reduction of MR inter-slice spacing. The SA-INR aims to represent an MR image as a continuous implicit function of 3D coordinates. In this way, the SA-INR can reconstruct the MR image with arbitrary inter-slice spacing by continuously sampling the coordinates in 3D space. In particular, a local-aware spatial attention operation is introduced to model nearby voxels and their affinity more accurately in a larger receptive field. Meanwhile, to improve the computational efficiency, a gradient-guided gating mask is proposed for applying the local-aware spatial attention to selected areas only. We evaluate our method on the public HCP-1200 dataset and the clinical knee MR dataset to demonstrate its superiority over other existing methods.