English

Adversarial and Perceptual Refinement for Compressed Sensing MRI Reconstruction

Computer Vision and Pattern Recognition 2018-07-02 v1

Abstract

Deep learning approaches have shown promising performance for compressed sensing-based Magnetic Resonance Imaging. While deep neural networks trained with mean squared error (MSE) loss functions can achieve high peak signal to noise ratio, the reconstructed images are often blurry and lack sharp details, especially for higher undersampling rates. Recently, adversarial and perceptual loss functions have been shown to achieve more visually appealing results. However, it remains an open question how to (1) optimally combine these loss functions with the MSE loss function and (2) evaluate such a perceptual enhancement. In this work, we propose a hybrid method, in which a visual refinement component is learnt on top of an MSE loss-based reconstruction network. In addition, we introduce a semantic interpretability score, measuring the visibility of the region of interest in both ground truth and reconstructed images, which allows us to objectively quantify the usefulness of the image quality for image post-processing and analysis. Applied on a large cardiac MRI dataset simulated with 8-fold undersampling, we demonstrate significant improvements (p<0.01p<0.01) over the state-of-the-art in both a human observer study and the semantic interpretability score.

Keywords

Cite

@article{arxiv.1806.11216,
  title  = {Adversarial and Perceptual Refinement for Compressed Sensing MRI Reconstruction},
  author = {Maximilian Seitzer and Guang Yang and Jo Schlemper and Ozan Oktay and Tobias Würfl and Vincent Christlein and Tom Wong and Raad Mohiaddin and David Firmin and Jennifer Keegan and Daniel Rueckert and Andreas Maier},
  journal= {arXiv preprint arXiv:1806.11216},
  year   = {2018}
}

Comments

To be published at MICCAI 2018

R2 v1 2026-06-23T02:45:31.776Z