High Frame Rate Ultrasound Particle Image Velocimetry for Estimating High Velocity Flow Patterns in the Left Ventricle

Abstract

Echocardiographic determination of multi-component blood flow dynamics in the left ventricle remains a challenge. In this study we compare contrast enhanced, high frame rate (1000 fps) echo particle image velocimetry (ePIV) against optical particle image velocimetry (oPIV, gold standard), in a realistic left ventricular phantom. We find that ePIV compares well to oPIV, even for the high velocity inflow jet (normalized RMSE = 9 ±1%). Additionally, we perform the method of Proper Orthogonal Decomposition, to better qualify and quantify the differences between the two modalities. We show that ePIV and oPIV resolve very similar flow structures, especially for the lowest order mode with a cosine similarity index of 86%. The coarser resolution of ePIV does result in increased variance and blurring of smaller flow structures when compared to oPIV. However, both modalities are in good agreement with each other for the modes that constitute the bulk of the kinetic energy. We conclude that high frame rate ePIV can accurately estimate the high velocity diastolic inflow jet and the high energy flow structures in a left ventricular setting.