Modification of the phased-tracking method for reduction of artifacts in estimated artery wall deformation

Noninvasive measurement of mechanical properties, such as elasticity, of the arterial wall, is useful for diagnosis of atherosclerosis. For assessment of mechanical properties, it is necessary to measure the deformation of the arterial wall. In this study, a modification of the previously proposed phased-tracking method was conducted to improve measurement of the small change in thickness (deformation) of the arterial wall due to the heartbeat. In our previous method, a set of two points along an ultrasonic beam was initially assigned, and the change in thickness of the layer between these two points during an entire cardiac cycle was estimated. In motion estimation with ultrasound, the motion of an interface or a scatterer, which generates an echo, can be obtained by estimating the change in time delay of the echo. For example, in the case of a carotid artery of a healthy subject, there are only two dominant echoes from the lumen-intima and media-adventitia interfaces. Thus, only the displacements of the lumen-intima and media-adventitia interfaces can be estimated, which means that ultrasound can estimate only the change in distance (thickness) between these two interfaces. However, even in this case, our previous method gives different estimates of the change in thickness, depending on the depths (positions in the arterial radial direction) of the two initially assigned points. In this study, modifications of the previous method in terms of the strategy for assignment of layers and the required thickness of an assigned layer were made to reduce such an artificial spatial variation in the estimated changes in thickness. Using the proposed method, errors in estimated changes in thickness were reduced from 21.2 plusmn 24.1% to 0.19 plusmn 0.04% (mean plusmn standard deviation) in simulation experiments. As in the case of the simulation experiments, the spatial variation in estimated changes in thickness also was reduced in in vivo experiments in a carotid artery of a healthy subject and in vitro experiments using two excised, diseased arteries