Combined acquisition of diffusion and T2*-weighted measurements using simultaneous multi-contrast magnetic resonance imaging

Abstract

Object: In this work, we present a technique called simultaneous multi-contrast imaging (SMC) to acquire multiple contrasts within a single measurement. Simultaneous multi-slice imaging (SMS) shortens scan time by allowing the repetition time (TR) to be reduced for a given number of slices. SMC imaging preserves TR, while combining different scan types into a single acquisition. This technique offers new opportunities in clinical protocols where examination time is a critical factor and multiple image contrasts must be acquired.

Materials and methods: High-resolution, navigator-corrected, diffusion-weighted imaging was performed simultaneously with T₂*-weighted acquisition at 3 T in a phantom and in five healthy subjects using an adapted readout-segmented EPI sequence (rs-EPI).

Results: The results demonstrated that simultaneous acquisition of two contrasts (here diffusion-weighted imaging and T₂*-weighting) with SMC imaging is feasible with robust separation of contrasts and minimal effect on image quality.

Discussion: The simultaneous acquisition of multiple contrasts reduces the overall examination time and there is an inherent registration between contrasts. By using the results of this study to control saturation effects in SMC, the method enables rapid acquisition of distortion-matched and well-registered diffusion-weighted and T₂*-weighted imaging, which could support rapid diagnosis and treatment of acute stroke.

Keywords: Diffusion; Echo-planar imaging; Stroke.

Fig. 1

Pulse sequence diagram of the rs-EPI sequence for SMC imaging. Data are acquired at two slice positions at the same time with one slice (A) generating DW-while the other slice (B) provides T₂*W image contrast. Firstly, slice A is excited and DW preparation is applied. Then slice B is excited before both signals are read out simultaneously using rs-EPI with a variable amplitude encoding gradient in the readout (G_RO) direction (labelled with arrows) and a blipped phase-encoding gradient (G_PE). A blipped-CAIPIRINHA gradient scheme along the slice-select (G_slice) direction is used in conjunction with receiver phase modulation to shift the T₂*W image by half a FOV in the phase-encoding direction. Finally, an RF refocusing pulse is applied to slice A only to generate a 2D navigator signal for phase correction. Note that a binomial water-excitation pulse was used in this study for the T₂*W excitation

Fig. 2

Data from five out of 20 measured SMC slices. The first row shows the collapsed SMC data with an in-plane acceleration factor of 2; The slices from the DW acquisition (with b = 0 s/mm²) are at the center of the FOV while the T₂*W slices are shifted by FOV/2 in the anterior–posterior direction. The middle row shows the DWI data and the bottom row shows the T₂*W data

Fig. 3

a Measured T₁ and B₁ maps in the spherical water-based phantom, b Simulated magnetization signal changes [%] with Bloch equations on the basis of the T₁ and B₁ map, c DW acquisition without and with simultaneous acquisition of T₂*W contrast, the corresponding difference image and signal-change map, d T₂*W contrast without and with simultaneous acquisition of the DW contrast, the corresponding difference image and signal-change map. The difference images show evidence of an uncorrected frequency drift during measurement in the outer ring. These areas of high differences are masked out in the saturation maps. Positive values [%] of signal change means smaller grey values in the SMC images than in the single-contrast images and vice versa

Fig. 4

Signal change result for slice iteration scheme A: a Measured T₁ and B₁ maps in vivo, b Simulated magnetization signal change [%] with Bloch equations on the basis of the T₁ and B₁ map in white matter, c DW acquisition without and with simultaneous acquisition of T₂*W contrast, the corresponding difference image and signal change map, d T₂*W contrast without and with simultaneous acquisition of the DW contrast, the corresponding difference image and signal change map. Positive values [%] of signal change means smaller grey values in the SMC images than in the single-contrast images and vice versa

Fig. 5

Signal change results for slice iteration scheme B: a Measured T₁ and B₁ maps in vivo, b Simulated magnetization signal change [%] with Bloch equations on the basis of the T₁ and B₁ map in white matter, c DW acquisition without and with simultaneous acquisition of T₂*W contrast, the corresponding difference image and signal change map, d T₂*W contrast without and with simultaneous acquisition of the DW contrast, the corresponding difference image and signal change map. Positive values [%] of signal change means smaller grey values in the SMC images than in the single-contrast images and vice versa

Fig. 6

a Comparison between simulated and measured signal change with and without split slice-GRAPPA reconstruction in the spherical phantom. b The table shows the mean signal change in terms of signal saturation and standard deviations for both contrasts in the phantom between three repeated measurements. c Comparison between simulated and measured signal change in vivo in grey and white matter (GM, WM) respectively

Fig. 7

a Difference images between rs-EPI single-contrast (DW trace-weighted and T₂*W) and rs-EPI SMC (separated trace-weighted and T₂*W contrast) in-vivo central slice out of 20 slices. b Histograms of all slices in all subjects of the masked difference image

Fig. 8

a The structural similarity index measure (SSIM) maps of the central slice of the calculated trace-weighted data for TR = 4500 ms and TR = 2250 ms. b Respective scatter plots of the gray level distributions of single-contrast and SMC data in the central slice

Fig. 9

Comparison between the registered TrW single-contrast and central SMC slices of subject 2 with the corresponding SSIM map for TR and TR/2. The close-up image allows better interpretation of the correlation between high and low SSIM values and structural changes between the single-contrast and SMC slices

Fig. 10

Contrast comparison in the central slice in one of four subjects. a Comparison of signal profile between single-contrast and SMC trace-weighted image. b Comparison of contrast between veins and tissue in T₂*W contrast

Fig. 11

Comparison between an T₂*-weighted SMC, 2D Flash and GRE-EPI (both product) measurement in terms of scan time and visual vein-tissue contrast