MR image super resolution by combining feature disentanglement CNNs and vision transformers

Dwarikanath Mahapatra, Zongyuan Ge

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

1 Citation (Scopus)

Abstract

State of the art magnetic resonance (MR) image super-resolution methods (ISR) using convolutional neural networks (CNNs) leverage limited contextual information due to the limited spatial coverage of CNNs. Vision transformers (ViT) learn better global context that is helpful in generating superior quality HR images. We combine local information of CNNs and global information from ViTs for image super resolution and output super resolved images that have superior quality than those produced by state of the art methods. We include extra constraints through multiple novel loss functions that preserve structure and texture information from the low resolution to high resolution images.

Original languageEnglish
Title of host publication5th International Conference on Medical Imaging with Deep Learning, MIDL 2022
EditorsEnder Konukoglu, Bjoern Menze, Archana Venkataraman, Christian F. Baumgartner, Qi Dou, Shadi Albarqouni
Place of PublicationLondon UK
PublisherProceedings of Machine Learning Research (PMLR)
Pages858-878
Number of pages21
Volume172
Publication statusPublished - 2022
EventInternational Conference on Medical Imaging with Deep Learning 2022 - Zurich, Switzerland
Duration: 6 Jul 20228 Jul 2022
Conference number: 5th
https://proceedings.mlr.press/v172/ (Published Proceedings)

Conference

ConferenceInternational Conference on Medical Imaging with Deep Learning 2022
Abbreviated titleMIDL 2022
Country/TerritorySwitzerland
CityZurich
Period6/07/228/07/22
Internet address

Keywords

  • CNN
  • disentanglement
  • MRI
  • super resolution
  • ViT

Cite this