Near-optimum carotid disease analysis using ultrasound image data

Abstract

The aim of this paper is to develop a new set-up to infer on significant disturbances in the downstream flow field in a stenosed carotid artery based on the analysis of clinical ultrasound acquisitions. A patient-specific numerical simulation system for the analysis of arterial blood flow under pulsatile conditions using an artificial neural network (ANN) with optimal configuration was implemented. The ANN topology was regarded as a solution of an optimization problem learning from carotid ultrasound data imaging. An iterative process based on a genetic search found the optimal ANN design, namely the number of nodes of the hidden layer, the weights and the bias of neurons and activation functions. The ANN predictions of patient-specific axial velocity patterns were introduced as boundary conditions for carotid flow simulation under pulsatile conditions. The hemodynamic behaviour of a stenotic carotid artery is addressed. © Civil-Comp Press, 2014.