Improved Techniques for Statistical Analysis of the Envelope of Backscattered Ultrasound Using the Homodyned K Distribution

Biomedical ultrasound is a prominent imaging modality for diagnostics. Conventional ultrasonic imaging is qualitative in nature with spatial resolution up to hundreds of micrometers. Quantitative ultrasound techniques based on ultrasonic backscatter can provide estimates describing tissue microstructure. Improving quantitative ultrasound techniques will result in improved diagnostic capabilities of ultrasound. Quantitative techniques were developed and assessed based on the envelope of backscattered ultrasound. The envelope of backscattered ultrasound can be modeled as the superposition of the scattered signals from individual scatterers in the medium being interrogated. As such, the envelope signal is statistical in nature. By applying a model to the amplitude distribution of the envelope, information about the sub-resolution material properties such as the scatterer number density and organizational structure can be obtained. The homodyned K distribution was used to model the envelope of backscattered ultrasound. An efficient parameter estimation algorithm was developed and tested through simulations and experiments. Techniques to reduce estimate bias and variance were assessed. The diagnostic potential of tissue characterization based on envelope statistics was evaluated.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Electrical and Computer Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2009

Author dc:creator

• Hruska, David P.

Contributors dc:contributor

• Oelze, Michael L.

• Ultrasonic tissue characterization
• ultrasound backscatter
• envelope statistics
• homodyned K distribution

Statement dc:rights

• Copyright 2009 David P. Hruska

Handle dc:identifier

http://hdl.handle.net/2142/11984

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/11984