A Deep Cascade of Convolutional Neural Networks for MR Image Reconstruction

By Jo Schlemper, Jose Caballero, Joseph V. Hajnal, Anthony N. Price, Daniel Rueckert. Information Processing in Medical Imaging, 2017

370 citations recorded by the source corpus.

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.

Subfields: Computational Mechanics, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/66c17841c9e5b648f4ec14828e1fdcb328157cc5

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

fastMRI: An Open Dataset and Benchmarks for Accelerated MRI (2018, background): “Research using small-scale proprietary raw k-space datasets is also common [15, 36, 35, 33, 22].”
Deep-learning inversion: a next generation seismic velocity-model building method (2019, background): “Many DL algorithms are built with CNNs and provide state-of-the-art performance in challenging inverse problems such as image reconstruction (Schlemper et al., 2017), superresolution (Dong et al.”
Neural Proximal Gradient Descent for Compressive Imaging (2018, background): “There is a handful of recent attempts to integrate the priors offered by generative nets for inverting linear inverse tasks dealing with local image restoration such as superresolution [10, 12], inpainting [13]; and more global tasks such as biomedical image reconstruction [15, 16, 17, 19, 20, 21, 2”
Secure and Robust Machine Learning for Healthcare: A Survey (2020, background): “tion of lungs nodules [4], medical image reconstruction [5], [6], and brain tumor segmentation [7], to name a few.”
MoDL: Model-Based Deep Learning Architecture for Inverse Problems (2017, methodology): “End-to-end training schemes [18], [19] rely on similar architecture as pre-trained models, but the CNN parameters at different iterations are trained such that the resulting deep network recovers the images from undersampled measurements.”
Convolutional Recurrent Neural Networks for Dynamic MR Image Reconstruction (2017, positive): “[18] proposed a cascade network which simulated the iterative reconstruction of dictionary learning-based methods and were later extended for dynamic MR reconstructions [3].”
ADMM-CSNet: A Deep Learning Approach for Image Compressive Sensing (2020, methodology): “For arbitrary-sizedwhole image reconstruction, some methods [45], [46], [47], [48], [49] have been introduced to use convolutional layers to learn a mapping from preliminary reconstructed results (e.”
CNN-Based Projected Gradient Descent for Consistent CT Image Reconstruction (2017, background): “Recently, in [25], a cascade of CNNs is used to reconstruct images.”
Deep Generative Adversarial Neural Networks for Compressive Sensing (GANCS) MRI (2018, background): “in [26] for Cartesian undersampling of 2D cardiac MR images”
On instabilities of deep learning in image reconstruction and the potential costs of AI (2019, methodology): “Deep MRI [10]: This neural network is for medium resolution single coil MRI with 33% subsampling.”