ERNAS: An Evolutionary Neural Architecture Search for Magnetic Resonance Image Reconstructions

TL;DR

This paper introduces ERNAS, an evolutionary neural architecture search method that optimizes deep learning models for MRI reconstruction, significantly improving efficiency and quality over manually designed networks.

Contribution

The paper presents a novel evolutionary neural architecture search algorithm tailored for MRI reconstruction, outperforming existing manually designed neural networks.

Findings

ERNAS outperforms manual models in MRI reconstruction quality

Optimized architectures improve reconstruction speed and accuracy

Results validated on brain and knee MRI datasets

Abstract

Magnetic resonance imaging (MRI) is one of the noninvasive imaging modalities that can produce high-quality images. However, the scan procedure is relatively slow, which causes patient discomfort and motion artifacts in images. Accelerating MRI hardware is constrained by physical and physiological limitations. A popular alternative approach to accelerated MRI is to undersample the k-space data. While undersampling speeds up the scan procedure, it generates artifacts in the images, and advanced reconstruction algorithms are needed to produce artifact-free images. Recently deep learning has emerged as a promising MRI reconstruction method to address this problem. However, straightforward adoption of the existing deep learning neural network architectures in MRI reconstructions is not usually optimal in terms of efficiency and reconstruction quality. In this work, MRI reconstruction from undersampled data was carried out using an optimized neural network using a novel evolutionary neural architecture search algorithm. Brain and knee MRI datasets show that the proposed algorithm outperforms manually designed neural network-based MR reconstruction models.