P5C-2 A New Convolution-Based Methodology to Simulate Ultrasound Images in a 2D / 3D Sector Format

Abstract

A study is presented in which a new convolution-based methodology has been developed to simulate ultrasound images in a 2D/3D sector format. In order to do so, the datasets were produced by reducing the conventional 2D/3D convolution model into multiple 1D convolutions (one for each line). After scan conversion of the individual image lines, conventional grayscale images were obtained. In this manuscript, a 2D / 3D image of a virtual left ventricle was simulated as an example. Moreover, the grayscale histogram statistics were verified to be Rayleigh distributed.