Possible detection of turbulent blood flow using multiparametric encoding gradients in MRI

D. Kwiat, S. Einav, D. Elad

Research output: Contribution to journalLetterpeer-review

Abstract

Magnetic resonance imaging (MRI) has significant potential as a highly accurate noninvasive flow measurement technique. Presented here is an approach to the imaging of turbulence in the velocity profiles. Recent publications have presented multiparametric encoding gradient methods, which are based on a deterministic rather than a statistical approach to spin motion. Following these methods of building up a general gradient from a basic set of gradients, the effects of turbulence and undetermined moments of motion on the image are discussed. Image blurring due to the nondeterministic behavior of spins (e.g., diffusion, turbulence) is not removed by these techniques, and this fact may be useful in identifying regions of turbulent flow, which are of importance in the clincial observation of cardiac and vascular haemodynamics. Due to its random nature, the elimination of turbulent effects by the use of a deterministic method is bound to fail. On the other hand, regions of turbulent flow may be identified due to signal decrease from those regions, provided one is careful to remove all the causes of nonturbulent signal reduction present due to nonturbulent flow moments. In the multiparametric phase encoding method, gradient amplitude tends to increase with higher moments of motion. The temporal behavior of these gradients is discussed, and it is suggested that the increase in their duration will enhance the encoding of higher motion terms at the expense of imaging time.

Original languageEnglish
Pages (from-to)316-323
Number of pages8
JournalMedical Physics
Volume18
Issue number2
DOIs
StatePublished - Mar 1991

Fingerprint

Dive into the research topics of 'Possible detection of turbulent blood flow using multiparametric encoding gradients in MRI'. Together they form a unique fingerprint.

Cite this