A Composite Statistical Model for Ultrasound Applications

Sevan Harput, Paschalis Sofotasios, Steven Freear

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Abstract

The statistics of the ultrasound echo contains information regarding the scattering properties of the tissue. The tissue characterization can be carried out through the ultrasound scattering by using statistical models. By matching the probability density function (PDF) of statistical distributions to the envelope of backscattered echo, the ultrasonic tissue characterization is possible. However, the distribution of scatterers in biological tissue does not simply obey the Rayleigh statistical models. Thus, it is significant to harness the pre-Rayleigh and post-Rayleigh statistics of backscattered signals. Nakagami-m/lognormal distribution has been considered one of the optimum statistical distribution to capture echo envelope statistics. However, due to the complexity of the algebraic representation of its PDF its use has been largely limited. In this study, an explicit expression for the PDF of Nakagami-m/lognormal distribution is derived. The validity of the results is verified through comparisons with results obtained from numerical integrations while its usefulness in ultrasound is shown through comparisons with realistic simulated ultrasound scenarios.
Original languageEnglish
Title of host publication2011 IEEE International Ultrasonics Symposium
DOIs
Publication statusPublished - 2011
Externally publishedYes
Publication typeA4 Article in a conference publication

Fingerprint

Dive into the research topics of 'A Composite Statistical Model for Ultrasound Applications'. Together they form a unique fingerprint.

Cite this