Premium
Effects of region‐of‐interest length on estimates of myocardial ultrasonic attenuation and backscatter
Author(s) -
Baldwin Steven L.,
Holland Mark R.,
Sosnovik David E.,
Miller James G.
Publication year - 2005
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.1852793
Subject(s) - region of interest , backscatter (email) , attenuation , scanner , receiver operating characteristic , biomedical engineering , materials science , length measurement , ultrasonic sensor , acoustics , optics , computer science , mathematics , medicine , physics , artificial intelligence , statistics , telecommunications , wireless
Measurements of tissue properties using an image‐based technique that makes use of an external reference may have the potential for practical clinical implementation in echocardiography. The objective of this study was to quantify the ability of this technique to distinguish myocardial attenuation and backscatter properties for specific lengths of the region‐of‐interest (ROI). We chose to exploit the anisotropic properties of the myocardium as a model for distinguishing tissue with different acoustic properties. Excised lateral wall segments from seven healthy adult sheep hearts were imaged using a commercially available (Philips/ATL) clinical scanner operating in the fundamental imaging mode with a linear array (L 7‐4). Statistical and receiver operating characteristic (ROC) analyses were used to evaluate the ability of the video signal analysis method to differentiate midmyocardial from subendocardial regions based on measurements of the acoustic properties for specific lengths of the ROI. Results demonstrate that the ability to distinguish tissue properties increases with ROI length for both slope of attenuation and backscatter coefficient measurements. Statistically significant differences were observed for measurements utilizing the ROI lengths as short as 0.4 cm with corresponding progressively increasing areas under the ROC curves for increasing ROI lengths. [NIH R37 HL40302]