Robust abdominal imaging with incomplete breath-holds

Nadine Gdaniec*, Holger Eggers, Peter Börnert, Mariya Doneva, Alfred Mertins

*Corresponding author for this work
9 Citations (Scopus)


Purpose Breath-holding is an established strategy for reducing motion artifacts in abdominal imaging. However, the breath-holding capabilities of patients are often overstrained by scans with large coverage and high resolution. In this work, a new strategy for coping with resulting incomplete breath-holds in abdominal imaging is suggested. Methods A sampling pattern is designed to support image reconstruction from undersampled data acquired up to any point in time using compressed sensing and parallel imaging. In combination with a navigator-based detection of the onset of respiration, it allows scan termination and thus reconstruction only from consistent data, which suppresses motion artifacts. The spatial resolution is restricted by a lower bound of the sampling density and is increased over the scan, to strike a compromise with the signal-to-noise ratio and undersampling artifacts for any breath-hold duration. Results The sampling pattern is optimized in phantom experiments and is successfully applied in abdominal gradient-echo imaging including water-fat separation on volunteers. Conclusions The new strategy provides images in which motion artifacts are minimized independent of the breath-holding capabilities of patients, and which enhance in terms of spatial resolution, signal-to-noise ratio, and undersampling artifacts with the a priori unknown breath-hold duration actually achieved in a particular scan.

Original languageEnglish
JournalMagnetic Resonance in Medicine
Issue number5
Pages (from-to)1733-1742
Number of pages10
Publication statusPublished - 01.01.2014


Dive into the research topics of 'Robust abdominal imaging with incomplete breath-holds'. Together they form a unique fingerprint.

Cite this