Denoising Simulated Low-Field MRI (70mT) using Denoising Autoencoders (DAE) and Cycle-Consistent Generative Adversarial Networks (Cycle-GAN)

Fernando Vega; Abdoljalil Addeh; M. Ethan MacDonald

Denoising Simulated Low-Field MRI (70mT) using Denoising Autoencoders (DAE) and Cycle-Consistent Generative Adversarial Networks (Cycle-GAN)

Image and Video Processing 2023-07-14 v1 Computer Vision and Pattern Recognition

Authors: Fernando Vega , Abdoljalil Addeh , M. Ethan MacDonald

Abstract

In this work, a denoising Cycle-GAN (Cycle Consistent Generative Adversarial Network) is implemented to yield high-field, high resolution, high signal-to-noise ratio (SNR) Magnetic Resonance Imaging (MRI) images from simulated low-field, low resolution, low SNR MRI images. Resampling and additive Rician noise were used to simulate low-field MRI. Images were utilized to train a Denoising Autoencoder (DAE) and a Cycle-GAN, with paired and unpaired cases. Both networks were evaluated using SSIM and PSNR image quality metrics. This work demonstrates the use of a generative deep learning model that can outperform classical DAEs to improve low-field MRI images and does not require image pairs.

Keywords

low-field MRI denoising denoising autoencoder cycle-gan image quality simulated data

Cite

@article{arxiv.2307.06338,
  title  = {Denoising Simulated Low-Field MRI (70mT) using Denoising Autoencoders (DAE) and Cycle-Consistent Generative Adversarial Networks (Cycle-GAN)},
  author = {Fernando Vega and Abdoljalil Addeh and M. Ethan MacDonald},
  journal= {arXiv preprint arXiv:2307.06338},
  year   = {2023}
}

Comments

International Society of Magnetic Resonance in Medicine (ISMRM) 2023, Abstract Number 1764

Denoising Simulated Low-Field MRI (70mT) using Denoising Autoencoders (DAE) and Cycle-Consistent Generative Adversarial Networks (Cycle-GAN)

Abstract

Keywords

Cite

Comments

Related papers