Publication detail

Unrolled L+S decomposition for compressed sensing in magnetic resonance imaging

MOKRÝ, O. VITOUŠ, J.

Original Title

Unrolled L+S decomposition for compressed sensing in magnetic resonance imaging

Type

journal article - other

Language

English

Original Abstract

A deep unrolled reconstruction method for dynamic magnetic resonance imaging is developed, based on the low-rank + sparse model. A standard solver of this model is enriched with trainable structures, forming a deep neural network, and several variants of the unrolled algorithm are trained on a simulated dataset. Evaluation against the standard solver for the model shows improvement in terms of mean squared error with the same computational cost.

Keywords

magnetic resonance imaging, L+S model, reconstruction, unrolled optimization

Authors

MOKRÝ, O.; VITOUŠ, J.

Released

12. 12. 2022

Publisher

International Society for Science and Engineering, o.s.

Location

Brno

ISBN

1213-1539

Periodical

Elektrorevue - Internetový časopis (http://www.elektrorevue.cz)

Year of study

24

Number

3

State

Czech Republic

Pages from

86

Pages to

93

Pages count

8

URL

http://www.elektrorevue.cz/cz/clanky/zpracovani-signalu/0/unrolled-l-s-decomposition-for-compressed-sensing-in-magnetic-resonance-imaging--rozbaleny-l-s-rozklad-pro-komprimovane-snimani-pri-zobrazovani-magnetickou-rezonanci-/

BibTex

@article{BUT180329,
  author="Ondřej {Mokrý} and Jiří {Vitouš}",
  title="Unrolled L+S decomposition for compressed sensing in magnetic resonance imaging",
  journal="Elektrorevue - Internetový časopis (http://www.elektrorevue.cz)",
  year="2022",
  volume="24",
  number="3",
  pages="86--93",
  issn="1213-1539",
  url="http://www.elektrorevue.cz/cz/clanky/zpracovani-signalu/0/unrolled-l-s-decomposition-for-compressed-sensing-in-magnetic-resonance-imaging--rozbaleny-l-s-rozklad-pro-komprimovane-snimani-pri-zobrazovani-magnetickou-rezonanci-/"
}