Journal of Computer Science

Myocardial Motion Analysis of Echocardiography Images using Optical Flow Radial Direction Distribution

Slamet Riyadi, Mohd Marzuki Mustafa and Aini Hussain

DOI : 10.3844/jcssp.2011.1046.1051

Journal of Computer Science

Volume 7, Issue 7

Pages 1046-1051


Problem statement: Myocardial motion is important information for physicians in diagnosing cardiac abnormalities. The motion vector of myocardial can be computed using optical flow technique, which then can be further analyzed based on its magnitude and angle. In practice, physicians are not concern about the angle of vector itself, but are more interested on whether a segment is moving to the center or not. Approach: Therefore, in this study we propose a relative motion direction with respect to the center of the cardiac cavity, called radial direction, which is more useful for diagnosis. The radial direction is computed as the difference between the angle of optical flow at a point of interest and the angle between the point and the cavity center. Because of the difficulty in performing analysis based solely on individual vectors, it is helpful to visualize and extract the overall trend by representing motion vectors by their angular distribution. Results: This method has been tested on clinical echocardiography sequences and has been shown to be successful in providing a radial direction profile of every segment for each echocardiographic frame. A comparison between the normal angular distribution and the proposed radial direction profile was also presented. Conclusion: The proposed profile was shown to be successful in providing the pattern of segmental motion which is easier for physician to analyze the myocardial motion compared with the normal angular distribution as well as more invariant to segment locations.


© 2011 Slamet Riyadi, Mohd Marzuki Mustafa and Aini Hussain. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.