Computer Vision in Echocardiography: Observer-Independent, Autonomous Echo Analysis Using Wavelet Footprints
C.H. Jansen, M. Arigovindan, M. Sühling, M. Unser, S. Marsch, P. Hunziker
Proceedings of the Fifty-Second Annual Scientific Session of the American College of Cardiology (ACC'03), Chicago IL, USA, March 30-April 2, 2003, Journal of the American College of Cardiology, vol. 41, no. 6, supp. 2, pp. 538A, March 19, 2003.
Background: Image segmentation and feature extraction in dynamic ultrasound images is demanding because of the large noise content, the absence of continuous boundaries and the lack of constant intensities within objects. Therefore ultrasound imaging is largely subjective with high observer dependence.
Methods: Multidimensional, multiscale wavelet analysis produces highly specific “footprints” from echo images. We exploited this fact to create an analysis environment that is based on shape and motion specific wavelet subbands. To test the feasibility of multidimensional multiscale footprints we tried to identify the mitral valve autonomously in 182 cardiac ultrasound sequences. As gold standard for the correct location of the mitral valve in the cardiac ultrasound sequences valve localization was performed by an experienced echocardiographer.
Results: Correct location was found in 165 of 182 ultrasound sequences corresponding to 91% accuracy. The use of multidimensional multiscale footprints led to significant enhancement of the predefined structure (middle image) in the image and significantly suppressed noise and non related cardiac structures as shown in pixel statistics of the mitral valve region versus non related image parts (P < 0.05 by T Test).
Conclusion: Computer vision technology is applicable to clinical echo and allows autonomous localization of predefined cardiac structures transforming this subjective, qualitative art into a objective quantitative science.
@INPROCEEDINGS(http://bigwww.epfl.ch/publications/jansen0302.html,
AUTHOR="Jansen, C.H. and Arigovindan, M. and S{\"{u}}hling, M. and
Unser, M. and Marsch, S. and Hunziker, P.",
TITLE="Computer Vision in Echocardiography: {O}bserver-Independent,
Autonomous Echo Analysis Using Wavelet Footprints",
BOOKTITLE="Fifty-Second Annual Scientific Session of the American
College of Cardiology ({ACC'03})",
YEAR="2003",
editor="",
volume="41, no. 6, supp. 2, March 19, 2003",
series="Journal of the American College of Cardiology",
pages="538A",
address="Chicago IL, USA",
month="March 30-April 2,",
organization="",
publisher="",
note="")