A phantom study on ultrasonic measurement of arterial wall strain combined with tracking of translational motion

Abstract

Correlation-based techniques are often applied to ultrasonic rf echoes to obtain the arterial wall deformation (strain). In such methods, the displacement estimates are biased due to changes in center frequency of echoes. One of the reasons for the change in center frequency is the interference of echoes from scatterers within the wall. In the phased tracking method previously proposed for strain estimation by our group, the estimated displacement contains both the components due to the translational motion and strain. The translational motion is larger than strain by a factor of 10 and, thus, the error in the estimated displacement due to the change in center frequency mainly depends on translational motion and is often larger than the minute displacement due to strain. To reduce this error, in this study, a method is proposed in which the translational motion is compensated using the displacement of the luminal boundary estimated by the phased tracking method before correlating echoes between the frame before deformation and that at the maximum deformation to estimate the strain distribution within the wall. In basic experiments using phantoms made of silicone rubber, the estimation error was much reduced to 15.6% in comparison with 36.4% obtained by the previous method.