A Texture Matching Method Considering Geometric Transformations in Noninvasive Ultrasonic Measurement of Arterial Elasticity

Abstract

Measurement of arterial elasticity can provide an important reference for understanding arterial wall changes that may occur in the early stages of atherosclerosis. Conventional correlation-based methods for evaluating arterial wall movements consider only the translation, ignoring the rotation and deformation, which limits the accuracy of measurement of arterial displacement and its biomechanical properties. This article proposes a novel texture matching method based on ultrasonic B-mode image considering geometric transformations to accurately measure arterial displacement and acquire arterial elasticity noninvasively. The method was validated by simulated images with rotation and deformation and further by measurements in vitro arterial phantom and in vivo common carotid arteries of 20 healthy volunteers. Simulation results demonstrate that the method can improve the accuracy of measurement of arterial displacement. Experimental results show that the elastic modulus of the arterial phantom agrees well with the results obtained from mechanical tests, deviating only 4.1%. The mean elastic modulus of the common carotid arteries is 361.7 ± 93.5 kPa. The texture matching method was shown to be able to measure the displacement and elasticity of the arterial wall with complex geometric transformations and may be clinically useful for early detecting and monitoring atherosclerosis.