Geometric Approaches to Increase the Expressivity of Deep Neural Networks for MR Reconstruction

Abstract

Recently, deep learning approaches have been extensively investigated to reconstruct images from accelerated magnetic resonance image (MRI) acquisition. Although these approaches provide significant performance gain compared to compressed sensing MRI (CS-MRI), it is not clear how to choose a suitable network architecture to balance the trade-off between network complexity and performance. Recently, it was shown that an encoder-decoder convolutional neural network (CNN) can be interpreted as a piecewise linear basis-like representation, whose specific representation is determined by the ReLU activation patterns for a given input image. Thus, the expressivity or the representation power is determined by the number of piecewise linear regions. As an extension of this geometric understanding, this paper proposes a systematic geometric approach using bootstrapping and subnetwork aggregation using an attention module to increase the expressivity of the underlying neural network. Our method can be implemented in both k-space domain and image domain that can be trained in an end-to-end manner. Experimental results show that the proposed schemes significantly improve reconstruction performance with negligible complexity increases.

Accepted for IEEE JSTSP Special Issue on Domain Enriched Learning for Medical Imaging

References in corpus (15)

• Adam: A Method for Stochastic Optimization
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• TensorFlow: A system for large-scale machine learning
• Instance Normalization: The Missing Ingredient for Fast Stylization
• Deep Convolutional Neural Network for Inverse Problems in Imaging
• Squeeze-and-Excitation Networks
• Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)
• Image reconstruction by domain transform manifold learning
• Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks
• Compressed Sensing MRI Reconstruction using a Generative Adversarial Network with a Cyclic Loss
• fastMRI: An Open Dataset and Benchmarks for Accelerated MRI
• Deep Residual Learning for Compressed Sensing CT Reconstruction via Persistent Homology Analysis
• Deep Generative Adversarial Networks for Compressed Sensing Automates MRI
• Understanding Geometry of Encoder-Decoder CNNs
• Boosting CNN beyond Label in Inverse Problems

Cited by in corpus (4)

• Unsupervised MR Motion Artifact Deep Learning using Outlier-Rejecting Bootstrap Aggregation
• Deep learning for fast MR imaging: a review for learning reconstruction from incomplete k-space data
• Simultaneous super-resolution and motion artifact removal in diffusion-weighted MRI using unsupervised deep learning
• CycleQSM: Unsupervised QSM Deep Learning using Physics-Informed CycleGAN