Ultrasound RF signal, backscattered from an inhomogeneous parenchymal tissue has the character of a random signal. It is the random fluctuating nature of this signal that is responsible for the speckle pattern observed in the images. These random signals, nevertheless, bear information related to the random scattering structure of the medium. Statistical moments of the signal can serve as feature parameters that are related to mean scatterer spacing in the random structure. However, such a feature is biased because the statistical nature of the RF signal depends, not only on the random tissue scattering structure, but also on the resolution cell volume of the imaging system, which in turn depends on the center frequency fo and bandwidth df of the interrogating short pulse and beam pattern of the transducer. The major goal of this project was three fold. (1) To develop a 3 dimensional comprehensive model of the random tissue structure characterized by a mean scatterer spacing. The interscatterer spacing and scattering strengths were random variables with a defined probability density function. (2) To develop a procedure to simulate backscattered RF signal from the model in response to various interrogating ultrasound pulses. The simulation procedure takes into account the 3-D nature of the resolution cell volume and its effect on the RF signal. (3) To develop a procedure to extract statistical parameters from the signal as a function of resolution cell volume. A feature parameter that is extracted from the analysis is related to the mean scatterer spacing, and is independent of the imaging system's point spread function. The procedure was based on recent theoretical prediction by Sleefe et.al. and therefore implicitly this work has attempted to verify the theoretical results.
Library of Congress Subject Headings
Ultrasonic imaging--Mathematical models
Zhu, Hui, "Scatterer number density estimation for tissue characterization in ultrasound imaging" (1990). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus