Automated detection of arterial wall boundaries based on correlation between adjacent receive scan lines for elasticity imaging

In our series of studies on the ultrasonic assessment of the regional elasticity of the arterial wall, the change in the thickness of the arterial wall caused by the heartbeat was measured by the ultrasonic phased-tracking method. In the measurement of elasticity, the boundaries of the anterior and posterior walls must be assigned in advance. Currently the boundaries are manually determined by the operator, which is time-consuming and results in observer variability. In this paper, we propose an automated method for the detection of arterial wall boundaries using multiscale dynamic programming, in which the cost function includes the correlation term between ultrasonic echoes in adjacent receive scan lines. The correlation term enables boundary detection that is more robust than conventional methods against noises. The effectiveness of the proposed method was validated using a phantom and applied to the in vivo measurements of carotid arteries. The root mean square error between the results obtained by the proposed method and the manual assignment of the boundaries was significantly improved.