By Topic

Manufacturer and machine setting based variation of parameters resulting from model-based image processing of echocardiographic transmitral Doppler velocity profiles

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
A. F. Hall ; Cardiovascular Biophys. Lab., Washington Univ. Sch. of Med., St. Louis, MO, USA ; J. A. Bettlach ; S. P. Nudelman ; S. J. Kovacs

Model-based image processing (MBIP) of echocardiographic transmitral Doppler velocity profile (DVP) images eliminates the need for manual tracing of the contour or determination of its attributes by-eye. Quantitative characterization of diastolic function in terms of the model parameters (solution of the “inverse problem”) requires fitting a kinematic model of filling (the PDF formalism) to the maximum velocity envelope (MVE) extracted from the DVP image. Averaging the MVEs of several filling cycles (heart beats) prior to model fit greatly reduces the variance of the model parameter estimates. However, even with beat averaging, the MVE may depend on ultrasound machine type and the settings (e.g. gain, dynamic range, baseline filter, post processing, etc.) of the machine. To determine the manufacturer and machine setting based variation of the MVE, and model parameters (c,k) obtained from MBIP, a flow-phantom study utilizing constant and pulsatile flow was performed. Both numerical and flow-phantom experiments indicate that variations of c and k are a function of the magnitude of c and k and exhibit directional dependence (corresponding to a narrow valley in c-k space). A processing method which minimizes these effects was derived and tested. The results show that the variations in the size of the parameters (c,k) due to machine type and setting fall within c-k space domains of constant mean-square error. The variation due to machine type and setting is less than variation due to physiologic or inter-group variability. Hence, the authors' process is suitable for clinical application and for differentiation of normal vs. pathologic states within these limits

Published in:

Ultrasonics Symposium, 1996. Proceedings., 1996 IEEE  (Volume:2 )

Date of Conference:

3-6 Nov 1996