Can echocardiographic particle image velocimetry correctly detect motion patterns as they occur in blood inside heart chambers? A validation study using moving phantoms
Prinz, Christian Faludi, Reka Walker, Andrew Amzulescu, Mihaela Gao, Hang Uejima, Tokuhisa Fraser, Alan G Voigt, Jens-Uwe # ×
Cardiovascular Ultrasound vol:10
ABSTRACT: AIMS: To validate Echo Particle Image Velocimetry (PIV) METHODS: High fidelity string and rotating phantoms moving with different speed patterns were imaged with different high-end ultrasound systems at varying insonation angles and frame rates. Images were analyzed for velocity and direction and for complex motion patterns of blood flow with dedicated software. Post-processing was done with MATLAB-based tools (Dflow, JUV, University Leuven). RESULTS: Velocity estimation was accurate up to a velocity of 42 cm/s (r=0.99, p<0.001, mean difference 0.4+/-2 cm/s). Maximally detectable velocity, however, was strongly dependent on frame rate and insonation angle and reached 42 cm/s under optimal conditions. At higher velocities estimates became random. Direction estimates did depend less on velocity and were accurate in 80-90%. In-plane motion patterns were correctly identified with three ultrasound systems. CONCLUSION: Echo-PIV appears feasible. Velocity estimates are accurate, but the maximal detectable velocity depends strongly on acquisition parameters. Direction estimation works sufficiently, even at higher velocities. Echo-PIV appears to be a promising technical approach to investigate flow patterns by echocardiography.