fastMRI: An Open Dataset and Benchmarks for Accelerated MRI

TL;DR

The paper introduces the fastMRI dataset, a large-scale, publicly available collection of raw MRI data and images, aimed at advancing machine learning methods for accelerated MRI reconstruction.

Contribution

It provides a standardized dataset and evaluation framework to facilitate research in MRI reconstruction using machine learning, along with an educational introduction to MRI for ML researchers.

Findings

Established a large-scale MRI dataset for ML research

Created standardized benchmarks for MRI reconstruction

Enabled rapid progress in MRI image reconstruction techniques

Abstract

Accelerating Magnetic Resonance Imaging (MRI) by taking fewer measurements has the potential to reduce medical costs, minimize stress to patients and make MRI possible in applications where it is currently prohibitively slow or expensive. We introduce the fastMRI dataset, a large-scale collection of both raw MR measurements and clinical MR images, that can be used for training and evaluation of machine-learning approaches to MR image reconstruction. By introducing standardized evaluation criteria and a freely-accessible dataset, our goal is to help the community make rapid advances in the state of the art for MR image reconstruction. We also provide a self-contained introduction to MRI for machine learning researchers with no medical imaging background.