|
Activity Number:
|
522
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #305138 |
|
Title:
|
Exact Algorithms for Decompressing the Log-Compressed Echo Envelope of Medical Ultrasound Images
|
|
Author(s):
|
Kai-Sheng Song*+
|
|
Companies:
|
University of North Texas
|
|
Address:
|
Department of Mathematics, Denton, TX, 76203,
|
|
Keywords:
|
algorithm ; envelope distribution ; estimation ; log-compression ; ultrasound
|
|
Abstract:
|
In clinical ultrasonic imaging systems, B-mode images are displayed with the log-compressed backscattered envelope. Log-compression of the echo envelope is a standard procedure to reduce the dynamic range of B-scan images and enhance weak backscatterers. Unfortunately, it makes many existing parameter estimation procedures inapplicable because they require direct access to either the rf data or envelope values, which are not available in most commercial ultrasonic imaging systems. None of existing methods estimates the compression factor (CF) and the envelope distribution simultaneously. In this talk, we propose exact algorithms for decompressing the log-compressed echo envelope by simultaneous estimation of the CF and the distribution for the received echo envelope. Some numerical experiments are also provided to demonstrate the finite sample performance of the proposed method.
|
