Accelerated dual-venc 4D flow MRI with variable high-venc spatial resolution for neurovascular applications

Abstract

Purpose: Dual-velocity encoded (dual-venc or DV) 4D flow MRI achieves wide velocity dynamic range and velocity-to-noise ratio (VNR), enabling accurate neurovascular flow characterization. To reduce scan time, we present interleaved dual-venc 4D Flow with independently prescribed, prospectively undersampled spatial resolution of the high-venc (HV) acquisition: Variable Spatial Resolution Dual Venc (VSRDV).

Methods: A prototype VSRDV sequence was developed based on a Cartesian acquisition with eight-point phase encoding, combining PEAK-GRAPPA acceleration with zero-filling in phase and partition directions for HV. The VSRDV approach was optimized by varying z, the zero-filling fraction of HV relative to low-venc, between 0%-80% in vitro (realistic neurovascular model with pulsatile flow) and in vivo (n = 10 volunteers). Antialiasing precision, mean and peak velocity quantification accuracy, and test-retest reproducibility were assessed relative to reference images with equal-resolution HV and low venc (z = 0%).

Results: In vitro results for all z demonstrated an antialiasing true positive rate at least 95% for $R_{\text{PEAK}-\text{GRAPPA}}$ = 2 and 5, with no linear relationship to z (p = 0.62 and 0.13, respectively). Bland-Altman analysis for z = 20%, 40%, 60%, or 80% versus z = 0% in vitro and in vivo demonstrated no bias >1% of venc in mean or peak velocity values at any $R_{\mathrm{ZF}}$ . In vitro mean and peak velocity, and in vivo peak velocity, had limits of agreement within 15%.

Conclusion: VSRDV allows up to 34.8% scan time reduction compared to PEAK-GRAPPA accelerated DV 4D Flow MRI, enabling large spatial coverage and dynamic range while maintaining VNR and velocity measurement accuracy.

Keywords: 4D flow; acceleration; acquisition; dual-venc; neurovascular.

FIGURE 1

Schematic K‐space acquisition scheme for single timeframe in VSRDV 4D Flow MRI with z = 40% and R _PEAK‐_GRAPPA = 2 acceleration (A) and R _PEAK‐_GRAPPA = 5 acceleration (B), with points of k‐space indicated by black dots having LV and HV acquired, those with light gray dots having both LV and HV omitted in PEAK‐GRAPPA scheme, and blue dots denoting k‐space lines having LV acquired and HV zero‐filled. Zero‐filled K _y lines are alternated throughout the temporal domain (not shown). C, Sequence diagram showing interleaved eight‐point acquisition of LV and HV as in typical DV imaging

FIGURE 2

In vitro setup for neurovascular anatomical phantom with pulsatile flow (A), and detailed view of Circle of Willis region of the phantom (B)

FIGURE 3

A, In vitro DV reconstruction and data comparison schematic for antialiasing performance and hemodynamic quantification performance. B, In vivo DV reconstruction and volumetric registration process schematic. C, Post‐processing workflow for anatomical phantoms: velocity reconstruction, vessel segmentation, automated centerline extraction, and plane placement, plane‐wise flow, and velocity quantification

FIGURE 4

A, Representative images for in vitro data, including LV, HV, and Zero‐filled DV for $R_{\text{PEAK}-\text{GRAPPA}}$= 5 and z = 80%. Blue markers indicate regions of aliasing in LV and their correction in the DV images. B, Antialiasing true positive rate

FIGURE 5

For mean velocity (cm/s) at $R_{\text{PEAK}-\text{GRAPPA}}$ = 2 (A) and $R_{\text{PEAK}-\text{GRAPPA}}$ = 5 (B), and for peak velocity (cm/s) at $R_{\text{PEAK}-\text{GRAPPA}}$ = 2 (C) and $R_{\text{PEAK}-\text{GRAPPA}}$ 5 (D), all with z = 80%, Bland–Altman plots for all cross‐sectional plane locations in the entire anatomy, and in vitro FDNG showing the anatomical distribution of differences (Bland–Altman bias), between Reference and Zero‐Filled DV. Bias and offset are indicated on the Bland–Altman plot where statistically significant. Vessels are labeled in (A)

FIGURE 6

Time‐resolved values at multiple individual planes are demonstrated for in vivo measurements of mean velocity (A), and (B) peak velocity, both at $R_{\text{PEAK}-\text{GRAPPA}}$ = 5. Value of z is indicated by line color. Vessels are: RICA, right internal carotid artery; LICA, left internal carotid artery; RMCA, right middle cerebral artery; LMCA, right middle cerebral artery

FIGURE 7

For mean velocity (cm/s) at $R_{\text{PEAK}-\text{GRAPPA}}$ = 5, z = 40% (A) and $R_{\text{PEAK}-\text{GRAPPA}}$ = 5, z = 80% (B), and for peak velocity (cm/s) at $R_{\text{PEAK}-\text{GRAPPA}}$ = 5 z = 40% $R_{\text{PEAK}-\text{GRAPPA}}$ = 5, z = 80% (D). Bland–Altman plots for all cross‐sectional plane locations in the entire anatomy, and in vivo FDNG showing the anatomical distribution of differences (Bland–Altman bias), relative to $R_{\text{PEAK}-\text{GRAPPA}}$ = 5, z = 0%. Bias and offset are indicated on the Bland–Altman plot where statistically significant. Vessels are labeled in (A)