Automated Doppler gate placement and velocity calculation based on a vessel angle estimate

Abstract

Pulsed-wave Doppler (PWD) sonography allows the quantitative measurement of blood flow in vessels if the vessel's diameter and the angle (Doppler angle theta) between the ultrasound beam and the vessel's flow direction are known. In current clinical routine, these parameters are manually determined by the examiner. However the manual determination is time-consuming and a source of error in blood flow measurements because of inaccuracy and variability. To overcome these problems, we present methods for an automated Doppler gate placement and Doppler angle estimation based on image processing algorithms. The Doppler gate is determined by analyzing the intensity profile along the ultrasound beam. To calculate the Doppler angle, we present a multiscale approach which estimates the vessel's flow direction by principal component analysis. A first evaluation shows promising results. Our automated approach yields parameters which match the parameters determined by clinical experts very well.

A feasability study of color flow doppler vectorization for automated blood flow monitoring

“…To the best of our knowledge, all attempts to apply computer vision to Doppler signals were based on either spectral or long axis color flow Doppler, this work being the first to apply it to color flow Doppler using the short axis vessel profile [5,7]. However, mathematical models of Doppler flow estimation through normal vector computation have been explored in the past [8].…”