RODEO: Robust DE-aliasing autoencOder for Real-time Medical Image Reconstruction

TL;DR

This paper introduces RODEO, a neural network-based autoencoder for real-time medical image reconstruction that is faster than compressed sensing and maintains high image quality by using a robust l1-norm loss function.

Contribution

The paper proposes a novel autoencoder approach with a robust loss function for rapid medical image reconstruction, enabling real-time performance.

Findings

Reconstruction speed is significantly faster than traditional compressed sensing methods.

Image quality is maintained with only slight degradation compared to standard methods.

The robust l1-norm loss improves autoencoder performance for this application.

Abstract

In this work we address the problem of real-time dynamic medical MRI and X Ray CT image reconstruction from parsimonious samples Fourier frequency space for MRI and sinogram tomographic projections for CT. Today the de facto standard for such reconstruction is compressed sensing. CS produces high quality images (with minimal perceptual loss, but such reconstructions are time consuming, requiring solving a complex optimization problem. In this work we propose to learn the reconstruction from training samples using an autoencoder. Our work is based on the universal function approximation capacity of neural networks. The training time for the autoencoder is large, but is offline and hence does not affect performance during operation. During testing or operation, our method requires only a few matrix vector products and hence is significantly faster than CS based methods. In fact, it is fast enough for real-time reconstruction the images are reconstructed as fast as they are acquired with only slight degradation of image quality. However, in order to make the autoencoder suitable for our problem, we depart from the standard Euclidean norm cost function of autoencoders and use a robust l1-norm instead. The ensuing problem is solved using the Split Bregman method.