Joint reconstruction of PET + parallel-MRI in a Bayesian coupled-dictionary MRF framework

Viswanath P. Sudarshan, Kratika Gupta, Gary Egan, Zhaolin Chen, Suyash P. Awate

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

2 Citations (Scopus)


Simultaneous positron emission tomography (PET) and magnetic resonance imaging (MRI) provide complementary information about brain function and structure. Joint reconstruction of MRI and PET images can improve image quality in both modalities, potentially enabling faster MRI and lower-dose PET scans. Current methods for joint MRI-PET reconstruction use priors that model inter-modality dependencies in image gradients. Many methods also ignore the potential in parallel MRI for acceleration. In contrast, we combine accelerated parallel MRI with a joint MRI-PET patch-based dictionary model to infer higher-order dependencies across MRI and PET image neighborhoods. We propose a novel Bayesian framework for joint MRI-PET reconstruction. The results show that our method reconstructs images more accurately, in simulated and in vivo MRI-PET (in parallel MRI) cases, than the state of the art.

Original languageEnglish
Title of host publicationMedical Image Computing and Computer Assisted Intervention – MICCAI 2019
Subtitle of host publication22nd International Conference Shenzhen, China, October 13–17, 2019 Proceedings, Part III
EditorsDinggang Shen, Tianming Liu, Terry M. Peters, Lawrence H. Staib, Caroline Essert, Sean Zhou, Pew-Thian Yap, Ali Khan
Place of PublicationCham Switzerland
Number of pages9
ISBN (Electronic)9783030322489
ISBN (Print)9783030322472
Publication statusPublished - 2019
EventMedical Image Computing and Computer-Assisted Intervention 2019 - Shenzhen, China
Duration: 13 Oct 201917 Oct 2019
Conference number: 22nd

Publication series

NameLecture Notes in Computer Science
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


ConferenceMedical Image Computing and Computer-Assisted Intervention 2019
Abbreviated titleMICCAI 2019
Internet address


  • Bayesian Markov random field
  • Joint dictionary
  • Joint reconstruction
  • Parallel MRI
  • SParsity
  • Undersampled k-space

Cite this