Cardiac dynamics compensation for intravascular ultrasound sequences

Abstract

Intravascular ultrasound (IVUS) is a well-established imaging technique for the assessment of coronary atherosclerotic plaque. IVUS has the ability to identify the arterial wall morphology, fibrous plaque locations and thicknesses, and internal lumen area. However, the motion of the imaging catheter with respect to the coronary wall caused by cardiac contraction during image acquisition impairs the subsequent visualization and quantification of IVUS image pullbacks. In this study, we propose a method to compensate for cardiac dynamics in IVUS. Keeping the original sensor field center origin unchanged, the vessel geometric centers are first extracted based on lumen segmentation. Subsequently, the periodic fluctuations of the geometric center sequence are used to formulate a heartbeat motion filtering algorithm based on the comb filter. Cardiac dynamics are then compensated by effectively filtering heartbeat deflection from both translational and rotational motion components. We evaluated 35 in vivo IVUS pullbacks and the experimental results showed an average percentage of artifact suppression of 77.82(± 7.55) %, demonstrating the reliability of the presented method in clinical cases. The method eliminates the need for image center correction and helps to maintain the original geometry of the vessel axes. Compared with existing methods, the presented method can more simply extract the heartbeat frequency and effectively filter out the higher order harmonic components of the heartbeat, resulting in more accurate cardiac dynamics compensation.