Brain MRI Super Resolution Using 3D Deep Densely Connected Neural Networks

TL;DR

This paper presents a novel 3D densely connected neural network architecture for super-resolving brain MRI images, significantly improving high-resolution detail restoration from low-resolution inputs using deep learning.

Contribution

Introduction of the 3D Densely Connected Super-Resolution Network (DCSRN) for brain MRI super-resolution, outperforming existing methods on a large dataset.

Findings

DCSRN outperforms bicubic interpolation in restoring high-resolution MRI images.

The network achieves superior results compared to other deep learning super-resolution methods.

Experiments on 1,113 subjects validate the effectiveness of the proposed approach.

Abstract

Magnetic resonance image (MRI) in high spatial resolution provides detailed anatomical information and is often necessary for accurate quantitative analysis. However, high spatial resolution typically comes at the expense of longer scan time, less spatial coverage, and lower signal to noise ratio (SNR). Single Image Super-Resolution (SISR), a technique aimed to restore high-resolution (HR) details from one single low-resolution (LR) input image, has been improved dramatically by recent breakthroughs in deep learning. In this paper, we introduce a new neural network architecture, 3D Densely Connected Super-Resolution Networks (DCSRN) to restore HR features of structural brain MR images. Through experiments on a dataset with 1,113 subjects, we demonstrate that our network outperforms bicubic interpolation as well as other deep learning methods in restoring 4x resolution-reduced images.