Cardiac CINE imaging with IDEAL water-fat separation and steady-state free precession

Scott B. Reeder*, Michael Markl, Huanzhou Yu, Jeffrey C. Hellinger, Robert J. Herfkens, Norbert J. Pelc

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

56 Scopus citations

Abstract

Purpose: to decompose multicoil CINE steady-state free precession (SSFP) cardiac images acquired at short echo time (TE) increments into separate water and fat images, using an iterative least-squares "Dixon" (IDEAL) method. Materials and Methods: Multicoil CINE IDEAL-SSFP cardiac imaging was performed in three volunteers and 15 patients at 1.5 T. Results: Measurements of signal-to-noise ratio (SNR) matched theoretical expectations and were used to optimize acquisition parameters. TE increments of 0.9-1.0 msec permitted the use of repetition times (TRs) of 5 msec or less, and provided good SNR performance of the water-fat decomposition, while maintaining good image quality with a minimum of banding artifacts. Images from all studies were evaluated for fat separation and image quality by two experienced radiologists. Uniform fat separation and diagnostic image quality was achieved in all images from all studies. Examples from volunteers and patients are shown. Conclusion: Multicoil IDEAL-SSFP imaging can produce high quality CINE cardiac images with uniform water-fat separation, insensitive to Bo inhomogeneities. This approach provides a new method for reliable fat-suppression in cardiac imaging.

Original languageEnglish (US)
Pages (from-to)44-52
Number of pages9
JournalJournal of Magnetic Resonance Imaging
Volume22
Issue number1
DOIs
StatePublished - Jul 2005

Keywords

  • Cardiac imaging
  • IDEAL
  • Magnetic resonance
  • Steady-state free precession
  • Water-fat separation

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Fingerprint

Dive into the research topics of 'Cardiac CINE imaging with IDEAL water-fat separation and steady-state free precession'. Together they form a unique fingerprint.

Cite this