Three‐point dixon technique for true water/fat decomposition with B₀ inhomogeneity correction

Abstract

An enhancement to Dixon's technique is described which can provide error-free decomposition of water and fat proton images even in the presence of off-resonance conditions which result from susceptibility differences, demagnetization, or shim errors. The method uses three measurements with phase shifts of 0, pi, and -pi between the fat and water resonances. The additional information provided by the third measurement is used to calculate an image of the field inhomogeneity in addition to true water and fat images. The signal-to-noise ratio (SNR) in the decomposed images is equivalent to that of a 2.7 NEX acquisition (instead of 3 NEX), yielding an SNR imaging efficiency of 95%. In addition, the B0 image which is provided may have diagnostic value in its own right. Examples of head and body scans often portray surprisingly large B0 shifts near interfaces between air or bone and soft tissue.

Dixon techniques for water and fat imaging

“…Note that because of the absolute value operation, the solution given by Equations [6] and [7] is not affected by the error phase 0 , even though 0 is in general spatially variant. When is not zero, Equations [6] and [7] do not provide a clean water and fat separation and the wateronly image or the fat-only image will in general contain admixture of both water and fat. To remedy the problem, Dixon proposed to form two magnitude images of ͉S 0 ͉ and ͉S 1 ͉ before they are summed or subtracted as in Equations [6] and [7].…”

“…When is not zero, Equations [6] and [7] do not provide a clean water and fat separation and the wateronly image or the fat-only image will in general contain admixture of both water and fat. To remedy the problem, Dixon proposed to form two magnitude images of ͉S 0 ͉ and ͉S 1 ͉ before they are summed or subtracted as in Equations [6] and [7]. However, it is easy to see that the water-only image thus obtained is really just an image where every pixel contains the dominant signal of the corresponding pixel, which can be either water or fat, rather than only water.…”

Quantitative MRI for the assessment of bone structure and function

“…Sixteen gradient echoes were collected, eight each before and after the phase-reversal pulse, at TE ¼ 4.6, 6.9, 9.2, 11.5, 13.8, 18.4, 23.0 and 27.6 ms and 57.0, 61.6, 66.2, 70.8, 73.1, 75.4, 77.7 and 80 ms, the latter being collected during the RF echo. Fat and water signal amplitudes were computed from echoes 1-3 by three-point Dixon processing (71) and, from these, the volume fractions of the two constituents. Of the 139 subjects studied 33 were men, 106 women (mean age 62.4 AE 11.4 years) and subjects had BMD T-scores scores ranging from þ3 to À5.…”

Three‐dimensional fast spoiled gradient‐echo dual echo (3D‐FSPGR‐DE) with water reconstruction: Preliminary experience with a novel pulse sequence for gadolinium‐enhanced abdominal MR imaging

“…The two-point Dixon image reconstruction uses a novel phase correction algorithm proposed by Ma (11) and Ma et al (10,12) that can efficiently and robustly decompose water and fat images from a DE acquisition. By compensating for radio frequency (RF) and magnetic field inhomogeneities during image reconstruction, the multipoint Dixon technique provides more homogeneous suppression of fat signal (13)(14)(15)(16). With FSPGR-DE the operator can choose to reconstruct all four image types, including in-phase, opposed-phase, water, and fat images, or can elect to reconstruct only the water and fat images.…”