Accuracy and limitations of vector flow mapping: left ventricular phantom validation using stereo particle image velocimetory

Figures

• 

  Fig. 1. Schematic of VFM algorithm

• 

  Fig. 2. Experimental system: a Schematics of the experimental setup, which consists of an LV phantom, an ultrasound scanner, and a PIV. A pulse signal mimicking an R-wave generated by an activator synchronizes all three systems at 1 Hz. b LV phantom in air (left) and in PEG400 (right) and c stereo-PIV velocity mapping of LV phantom. Contour indicates through-plane velocity component v z. Time above each frame indicates duration after triggering pulse

• 

  Fig. 3. Comparison of 2D velocity fields. PIV, VFM, and their discrepancies (i.e., differences between PIV and VFM data) are recorded at different time frames. Background color indicates magnitude of 2D vectors. Dashed lines indicate wall boundaries acquired by PIV. Color scale, indicating magnitude of vector, is optimized for each time frame

• 

  Fig. 4. Statistical analysis: a Probability distribution function (pdf) of absolute velocity error, b transition in standard deviation of vector discrepancy over CFM velocity range, and c correlation between VFM and PIV vectors. All vector components are plotted in graphs at t = 0.1 s and 0.6 s for all frames. Velocity range is set to fit all plots for both time frames. Dotted lines indicate y = x. d Circulation mapped at t = 0.6 s. Square with arrowhead indicates size and direction of line integral applied to calculate maps of circulation. e Local extremum of circulation (LEC) has been plotted for time period to indicate transition pathway. VFM estimation data represents the average of three cycles. Each LEC for three cardiac cycles is mapped by colored dotted line to demonstrate reproducibility

• 

  Fig. 5. Comparison of wall velocities. a Horizontal component v x of velocity of right wall at y = 15 mm, where wall motion is largest, b 2D vector mapping by PIV and VFM, and PIV reconstruction results (PIV-based VFM) at t = 0.9 s, c v θ distribution along y-axis at x = 10 mm, and d intensity mapping of through-plane component of PIV vectors with overlay of 2D vectors (left), continuity equation (middle), and Δv x = v xVFM − v xPIV (right). Dashed lines indicate wall boundaries