Quantitative analysis of ultrasound image texture characteristics is a promising method of detecting altered tissue structure. However, it is known that there is significant variation in quantitative texture measures due to the position of the target in the ultrasound beam. In this study, the authors demonstrate that part of this artifactual variation is due to the conventional method of data analysis, which uses scan-converted data, i. e. data in rectangular coordinates. This artifactual regional variation can be markedly reduced by analyzing image texture using the polar coordinates in which the data were originally acquired.
|Title of host publication||Computers in Cardiology|
|Editors||Kenneth L. Ripley|
|Number of pages||4|
|Publication status||Published - 1 Dec 1984|
ASJC Scopus subject areas
- Computer Science Applications
- Cardiology and Cardiovascular Medicine