Clinical image quality assessment of accelerated magnetic resonance neuroimaging using compressed sensing

Samir D. Sharma, Caroline L. Fong, Brian S. Tzung, Meng Law, Krishna S. Nayak

Research output: Contribution to journalArticleResearchpeer-review

50 Citations (Scopus)

Abstract

Purpose: The aim of this study was to determine to what degree current compressed sensing methods are capable of accelerating clinical magnetic resonance neuroimaging sequences. Methods: Two 2-dimensional clinical sequences were chosen for this study because of their long scan times. A pilot study was used to establish the sampling scheme and regularization parameter needed in compressed sensing reconstruction. These findings were used in a subsequent blinded study in which images reconstructed using compressed sensing were evaluated by 2 board-certified neuroradiologists. Image quality was evaluated at up to 10 anatomical features. Results: The findings indicate that compressed sensing may provide 2-fold acceleration of certain clinical magnetic resonance neuroimaging sequences. A global ringing artifact and image blurring were identified as the 2 primary artifacts that would hinder the ability to confidently discern abnormality. Conclusions: Compressed sensing is able to moderately accelerate certain neuroimaging sequences without severe loss of clinically relevant information. For those sequences with coarser spatial resolution and/or at a higher acceleration factor, artifacts degrade the quality of the reconstructed image to a point where they are of little to no clinical value.

Original languageEnglish
Pages (from-to)638-645
Number of pages8
JournalInvestigative Radiology
Volume48
Issue number9
DOIs
Publication statusPublished - 1 Sep 2013
Externally publishedYes

Keywords

  • accelerated imaging
  • clinical quality assessment
  • Compressed sensing
  • magnetic resonance neuroimaging

Cite this