Densely connected wavelet-based autoencoder for MR image reconstruction

Amir Aghabiglou*, Ender M. Eksioglu

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Citations (Scopus)

Abstract

Recently, methods based on deep learning have been introduced to the literature as a solution for accelerating magnetic resonance imaging technique. However, Image reconstruction from subsampled data is an ill-posed problem. In the current study, the wavelet package has been applied to deep networks. The replacement of the conventional downsampling and upsampling layers with Discrete Wavelet Transform (DWT) and Inverse Wavelet Transform (IWT) improved the reconstruction results. Moreover, the consequence of this substitution has been investigated on potent densely connected deep networks. The proposed novelty resulted in promising performance improvement in MR Image reconstruction.

Original languageEnglish
Title of host publication2022 45th International Conference on Telecommunications and Signal Processing, TSP 2022
EditorsNorbert Herencsar
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages212-215
Number of pages4
ISBN (Electronic)9781665469487
DOIs
Publication statusPublished - 2022
Event45th International Conference on Telecommunications and Signal Processing, TSP 2022 - Virtual, Online, Czech Republic
Duration: 13 Jul 202215 Jul 2022

Publication series

Name2022 45th International Conference on Telecommunications and Signal Processing, TSP 2022

Conference

Conference45th International Conference on Telecommunications and Signal Processing, TSP 2022
Country/TerritoryCzech Republic
CityVirtual, Online
Period13/07/2215/07/22

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

Keywords

  • Deep learning
  • Densely Connected Residual Network
  • Magnetic resonance imaging
  • MR Image Reconstruction

Fingerprint

Dive into the research topics of 'Densely connected wavelet-based autoencoder for MR image reconstruction'. Together they form a unique fingerprint.

Cite this