Experimental comparison of methods for measuring backscatter coefficient using single element transducers

Abstract

The backscatter coefficient (BSC) has promise as a diagnostic aid. However, measurements of the BSC of soft-tissue mimicking materials have proven difficult; results on the same samples by various laboratories have up to two orders of magnitude difference. This study compares methods of data analysis using data acquired from the same samples using single element transducers, with a frequency range of 1 to 20 MHz and pressure focusing gains between 5 and 60. The samples consist of various concentrations of milk in agar with scattering from glass microspheres. Each method utilizes a reference spectrum from a planar reflector but differ in the diffraction and attenuation correction algorithms. Results from four methods of diffraction correction and three methods of attenuation correction are compared to each other and to theoretical predictions. Diffraction correction varies from no correction to numerical integration of the beam throughout the data acquisition region. Attenuation correction varies from limited correction for the attenuation up to the start of the echo acquisition window, to correcting for attenuation within a numerical integration of the beam profile. Results indicate the best agreements with theory are the methods that utilize the numerical integration of the beam profile.