Fundamental performance assessment of 2-D myocardial elastography in a phased-array configuration

Abstract

Two-dimensional myocardial elastography, an RF-based, speckle-tracking technique, uses 1-D cross-correlation and recorrelation methods in a 2-D search, and can estimate and image the 2-D transmural motion and deformation of the myocardium so as to characterize the cardiac function. Based on a 3-D finite-element (FE) canine left-ventricular model, a theoretical framework was previously developed by our group to evaluate the estimation quality of 2-D myocardial elastography using a linear array. In this paper, an ultrasound simulation program, Field II, was used to generate the RF signals of a model of the heart in a phased-array configuration and under 3-D motion conditions; thus simulating a standard echocardiography exam. The estimation method of 2-D myocardial elastography was adapted for use with such a configuration. All elastographic displacements and strains were found to be in good agreement with the FE solutions, as indicated by the mean absolute error (MAE) between the two. The classified first and second principal strains approximated the radial and circumferential strains, respectively, in the phased-array configuration. The results at different sonographic signal-tonoise ratios (SNRs) showed that the MAEs of the axial, lateral, radial, and circumferential strains remained relatively constant when the SNRs was equal to or higher than 20 dB. The MAEs of the strain estimation were not significantly affected when the acoustic attenuation was included in the simulations. A significantly reduced number of scatterers could be used to speed up the simulation, without sacrificing the estimation quality.The proposed framework can further be used to assess the estimation quality, explore the theoretical limitation and investigate the effects of various parameters in 2-D myocardial elastography under more realistic conditions. © 2009 IEEE.