Improved determination of spin density, T1 and T2 from a three-parameter fit to multiple-delay-multiple-echo (MDME) NMR images

T. E. Conturo, R. R. Price, A. H. Beth, M. R. Mitchell, C. L. Partain, A. E. James

Research output: Contribution to journalArticlepeer-review

13 Scopus citations

Abstract

A method is presented for simultaneously determining values of relative hydrogen spin density Nr, T1 and T2 from a single set of NMR image intensities acquired in a short imaging time. Present methods use separate acquisitions and data sets to determine all three parameters. In the method presented, multiple-echo data are collected at multiple delays in virtually the same imaging time used to obtain T1 and a T2-weighted Nr from a separate saturation recovery (SR) T1 measurement. All three parameters are then determined by a three-parameter fit of a derived signal intensity equation to these multiple-delay-multiple-echo (MDME) data. This provides an inherent correction of Nr for T1 and T2 weighting without the use of sequences with TD>5T1, and without further data collection for a separate T2 measurement. It also provides an effective reduction in the noise of the separate T2 measurement. A three-parameter fit to MDME data appears to be superior to the separate T 1 and T2 measurements currently used to determine all three parameters. Calculations performed on CrCl3 solutions produced T1 values from 21 ms to 3.4 s, T2 values from 6 to 714 ms, and standard errors as low as 0.33%, with a net imaging time of the order of that required for routine low-noise signal intensity imaging. The method could potentially be used in NMR spectroscopy to give similar benefits.

Original languageEnglish
Article number005
Pages (from-to)1361-1380
Number of pages20
JournalPhysics in medicine and biology
Volume31
Issue number12
DOIs
StatePublished - 1986

Fingerprint

Dive into the research topics of 'Improved determination of spin density, T1 and T2 from a three-parameter fit to multiple-delay-multiple-echo (MDME) NMR images'. Together they form a unique fingerprint.

Cite this