Multi‐gradient‐echo myelin water fraction imaging: Comparison to the multi‐echo‐spin‐echo technique

    loading  Checking for direct PDF access through Ovid


Myelin is a membrane structure present in the central and peripheral nervous systems that surrounds axons and serves as an insulating material, enabling efficient transmission of axon potentials 1. Quantitative MRI of myelin can provide valuable insight into myelin development and pathology. This is of particular interest in the case of myelin‐related disorders, such as multiple sclerosis, wherein demyelination and remyelination can occur. Quantitative MRI of myelin has been pursued for many years with a variety of techniques.
One approach for quantifying myelin is based on analysis of JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM1/v/2018-01-24T161827Z/r/image-png relaxation times. The multicomponent nature of transverse relaxation times in central and peripheral nervous tissue has been established in a number of studies 2 using nuclear magnetic resonance relaxation measurements. At least three distinct components have been detected in the central nervous system and attributed to (in order of increasing JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM2/v/2018-01-24T161827Z/r/image-png times): water trapped between the lipid bilayers of myelin, axoplasmic water, and extracellular water. MacKay et al. 7 introduced an in vivo technique based on a single‐slice multi‐echo spin‐echo (MESE) CPMG sequence 8 to acquire JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM3/v/2018-01-24T161827Z/r/image-png relaxation imaging data, with subsequent multicomponent analysis. In their data from human brains, two sizable JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM4/v/2018-01-24T161827Z/r/image-png components ( JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM5/v/2018-01-24T161827Z/r/image-png of 10–40 ms and JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM6/v/2018-01-24T161827Z/r/image-png of 70–100 ms) were identified and attributed to myelin water (MW) and to the combination of axonal water and extra‐cellular water (collectively referred to as intra/extracellular water) 9, respectively. The data were analyzed using a regularized non‐negative least squares (NNLS) algorithm 10 to fit a pseudo‐continuous JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM7/v/2018-01-24T161827Z/r/image-png distribution to the magnitude MESE data; moreover, the myelin water fraction (MWF) was defined as the ratio of the area under the MW peak to the total water signal. Strong correlations have been observed between the MWF and histological measures of myelin 11, validating the use of the MWF as a biomarker for myelin content in tissue.
Unfortunately, MWF imaging based on the aforementioned single‐slice MESE sequence is not clinically feasible because of its long scan time. However, this technique has recently been extended to 3D imaging by the use of a combined gradient and spin echo sequence 14. Multicomponent analysis of gradient and spin echo JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM8/v/2018-01-24T161827Z/r/image-png data is based on regularized NNLS with concurrent correction for stimulated echo contamination of the signal decay curve 15.
Alternative multi‐slice and 3D approaches to MWF imaging that have been proposed include: JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM9/v/2018-01-24T161827Z/r/image-png ‐prepared imaging 16, multicomponent driven‐equilibrium single‐pulse observation of JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM10/v/2018-01-24T161827Z/r/image-png and JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM11/v/2018-01-24T161827Z/r/image-png (mcDESPOT) 18, and MWF mapping based on multi‐gradient‐echo (MGRE) imaging with multicomponent analysis of the JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM12/v/2018-01-24T161827Z/r/image-png decay curve 19. MGRE imaging has the advantage of offering fast multi‐slice (2D) or 3D whole brain coverage, with low specific absorption rate, and short first echo time (TE1) and echo spacing. Two methods have been used to analyze magnitude MGRE data: regularized 22 NNLS fitting of a pseudo‐continuous JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM13/v/2018-01-24T161827Z/r/image-png distribution 21 and a three‐component model 23. In the latter, the three components are assumed to be MW, extra‐cellular water, and axonal water.
An important caveat to MGRE‐based MWF imaging is the presence of macroscopic magnetic field inhomogeneities ( JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM14/v/2018-01-24T161827Z/r/image-png ), which cause non‐exponential signal decay and, consequently, biases in the MWF. The effect of JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM15/v/2018-01-24T161827Z/r/image-png on the exponential signal decay has been described for 2D and 3D acquisitions 25. Typically, in 2D multi‐slice acquisitions, voxel sizes are larger in the slice‐select direction, making magnetic field gradients in the slice‐select direction ( JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM16/v/2018-01-24T161827Z/r/image-png ) the dominant source of signal loss and JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM17/v/2018-01-24T161827Z/r/image-png bias. The effect of JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM18/v/2018-01-24T161827Z/r/image-png on 2D MGRE signal decay has been modeled as a sinc function that is dependent on JOURNAL/mrim/04.02/01445475-201803000-00022/math_22MM19/v/2018-01-24T161827Z/r/image-png and the slice thickness 26.

Related Topics

    loading  Loading Related Articles