MoDL: Model-Based Deep Learning Architecture for Inverse Problems

By Hemant Kumar Aggarwal, Merry P. Mani, Mathews Jacob. 2017

1,267 citations recorded by the source corpus.

We introduce a model-based image reconstruction framework with a convolution neural network (CNN)-based regularization prior. The proposed formulation provides a systematic approach for deriving deep architectures for inverse problems with the arbitrary structure. Since the forward model is explicitly accounted for, a smaller network with fewer parameters is sufficient to capture the image information compared to direct inversion approaches. Thus, reducing the demand for training data and training time. Since we rely on end-to-end training with weight sharing across iterations, the CNN weights are customized to the forward model, thus offering improved performance over approaches that rely on pre-trained denoisers. Our experiments show that the decoupling of the number of iterations from the network complexity offered by this approach provides benefits, including lower demand for training data, reduced risk of overfitting, and implementations with significantly reduced memory footprint. We propose to enforce data-consistency by using numerical optimization blocks, such as conjugate gradients algorithm within the network. This approach offers faster convergence per iteration, compared to methods that rely on proximal gradients steps to enforce data consistency. Our experiments show that the faster convergence translates to improved performance, primarily when the available GPU memory restricts the number of iterations.

Subfields: Biomedical Engineering, Computational Mechanics, Mathematical Physics

How researchers see this paper

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

In Biomedical Engineering

In Computational Mechanics

In Mathematical Physics

Citations in the corpus

An overview of deep learning in medical imaging focusing on MRI (2018, background): “(30)See [154, 155, 156, 157, 158] for recent perspectives and developments connecting deep learning-based reconstruction methods to the more general research field of inverse problems.”
Deep-Learning Methods for Parallel Magnetic Resonance Imaging Reconstruction: A Survey of the Current Approaches, Trends, and Issues (2019, methodology): “[25] is an example for such an approach.”
NETT: solving inverse problems with deep neural networks (2018, methodology): “In [2], the norm of a residual network is used as a regularizer.”
Deep Learning Techniques for Inverse Problems in Imaging (2020, methodology): “ting [105], [106], block coordinate descent [107]–[109], alternating minimization [82], and approximate message passing [88], [110].”
FISTA-Net: Learning a Fast Iterative Shrinkage Thresholding Network for Inverse Problems in Imaging (2020, methodology): “proposed MoDL, a variational framework involving a data-consistency term and a learned CNN to capture the image redundancy [26].”
Convolutional Recurrent Neural Networks for Dynamic MR Image Reconstruction (2017, positive): “More recently, several methods on 2D MR image reconstruction were proposed [9], [20], [21], which share similar idea with our proposed method that integrates data fidelity term and regularisation term into a single deep network so”
Deep Magnetic Resonance Image Reconstruction: Inverse Problems Meet Neural Networks (2020, neutral): “Another method, modelbased deep learning [8], solves the DC subproblem (18) using the conjugate gradient (CG) algorithm to handle more complex forward models (e.”