A Deep Cascade of Convolutional Neural Networks for MR Image Reconstruction (2017-03-01T00:00:00.000000Z)

TL;DR

This work proposes a framework for reconstructing MR images from undersampled data using a deep cascade of convolutional neural networks to accelerate the data acquisition process and shows that for Cartesian undersampling of 2D cardiac MR images, the proposed method outperforms the state-of-the-art compressed sensing approaches.

Abstract

The acquisition of Magnetic Resonance Imaging (MRI) is inherently slow. Inspired by recent advances in deep learning, we propose a framework for reconstructing MR images from undersampled data using a deep cascade of convolutional neural networks to accelerate the data acquisition process. We show that for Cartesian undersampling of 2D cardiac MR images, the proposed method outperforms the state-of-the-art compressed sensing approaches, such as dictionary learning-based MRI (DLMRI) reconstruction, in terms of reconstruction error, perceptual quality and reconstruction speed for both 3-fold and 6-fold undersampling. Compared to DLMRI, the error produced by the method proposed is approximately twice as small, allowing to preserve anatomical structures more faithfully. Using our method, each image can be reconstructed in 23 ms, which is fast enough to enable real-time applications.

Authors

D. Rueckert

14 papers

Jose Caballero

4 papers

Jo Schlemper

5 papers

TL;DR

Abstract

Authors

References37 items

Deep ADMM-Net for Compressive Sensing MRI

Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network

ReconNet: Non-Iterative Reconstruction of Images from Compressively Sensed Measurements

Compressed sensing reconstruction of dynamic contrast enhanced MRI using GPU-accelerated convolutional sparse coding

Accelerating magnetic resonance imaging via deep learning

Deep Residual Learning for Image Recognition

Multi-Scale Context Aggregation by Dilated Convolutions

MR Image Reconstruction with Convolutional Characteristic Constraint (CoCCo)

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

Compressed sensing MRI using sparsity induced from adjacent slice similarity

U-Net: Convolutional Networks for Biomedical Image Segmentation

A novel k-space annihilating filter method for unification between compressed sensing and parallel MRI

Low‐rank plus sparse matrix decomposition for accelerated dynamic MRI with separation of background and dynamic components

Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification

Image Super-Resolution Using Deep Convolutional Networks

Adam: A Method for Stochastic Optimization

Fast multi-contrast MRI reconstruction.

Going deeper with convolutions

ESPIRiT—an eigenvalue approach to autocalibrating parallel MRI: Where SENSE meets GRAPPA

Dictionary Learning and Time Sparsity for Dynamic MR Data Reconstruction

Learning Algorithms in Complex-Valued Neural Networks using Wirtinger Calculus

MR Image Reconstruction From Highly Undersampled k-Space Data by Dictionary Learning

Accelerated Dynamic MRI Exploiting Sparsity and Low-Rank Structure: k-t SLR

Magnetic resonance imaging in real time: Advances using radial FLASH

Accelerating SENSE using compressed sensing

Compressed Sensing MRI

On NUFFT-based gridding for non-Cartesian MRI.

Undersampled radial MRI with multiple coils. Iterative image reconstruction using a total variation constraint

Improved k–t BLAST and k–t SENSE using FOCUSS

Pattern Recognition and Machine Learning

Analysis and compensation of eddy currents in balanced SSFP

Best practices for convolutional neural networks applied to visual document analysis

SENSE: Sensitivity encoding for fast MRI

Continuous radial data acquisition for dynamic MRI

Learning representations by back-propagating errors

Compressed Sensing

Faster Imaging with Randomly Perturbed, Undersampled Spirals and |L|_1 Reconstruction

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names