Aliasing-Free Maximum Velocity Estimation using Multi-angle Plane Wave Vector Flow Imaging

Abstract

Maximum velocity indices such as peak systolic velocity (PSV), end-diastolic velocity (EDV), and pulsatility index (PI) are important biomarkers used within clinics to determine the degree of stenosis within arteries including the internal carotid artery and femoral arteries. These indices are also used within cardiovascular and cerebrovascular health research. Currently, maximum velocity indices within clinics and research are typically measured using conventional spectral doppler ultrasound, which requires the operator to manually move a small sample volume to the relevant area and manually adjust a doppler angle. As a result, maximum velocity biomarkers can only be measured within a small portion of the ultrasound region at one time. Further, operator errors and using a single doppler angle throughout the cardiac cycle is a known source of error.

High-frame rate ultrasound (HIFRUS) is a novel imaging modality that insonifies the entire ultrasound region simultaneously and is capable of capture flow velocities at thousands of frames per second. Instead of a small sample volume that measures maximum velocity biomarkers within a limited region, HIFRUS allows simultaneous derivation of velocity estimates across the entire ultrasound region.

The objective of this thesis is to develop a novel framework that obtains angle-independent maximum vector velocity estimates across the entire ultrasound region with sub-millisecond resolution. This framework incorporates an inexpensive IPS-based maximum velocity estimator and de-aliasing techniques with plane-wave imaging principles to obtain accurate maximum velocity estimates. To validate and evaluate this framework, three different phantom experiments are conducted. Maximum velocity estimates obtained within phantoms are quantitatively compared with values obtained within a clinical scanner. To evaluate the feasibility of this framework in-vivo, two datasets of the carotid and femoral bifurcation are imaged and evaluated.

By obtaining angle-independent vector velocity estimates throughout the entire ultrasound region at thousands of frames per second, our framework seeks to solve many issues that exist within conventional ultrasound scanners.