Regional Classification of Left Ventricular Wall in Small Animal Ultrasound Imaging
Heart diseases such as acute myocardial infarction are one of the main death causes in industrial societies today. In general, these diseases are accompanied by physiological changes in the heart, which can give valuable information for future therapy if early recognized by cardiologists. This work proposes a processing pipeline for classification of left ventricle regions in medical ultrasound images of small animals as a first step towards recognition of heart remodeling processes. Based on state-of-the-art methods from computer vision an automatic classification of image regions in healthy and scarred myocardial tissue is realized. The performance of the proposed pipeline is evaluated on real ultrasound data of living mice before and after artificially induced myocardial infarction.
KeywordsClassification tissue characterization high-level segmentation shape prior motion estimation optical flow echocardiography ultrasound imaging
Unable to display preview. Download preview PDF.
- 2.Duda, R.O., Hart, P.E., Stork, D.G.: Pattern Classification, 2nd edn. Wiley (2001)Google Scholar
- 3.Mahmood, S.S., Levy, D., Vasan, R.S., Wang, T.J.: The Framingham Heart Study and the Epidemiology of Cardiovascular Disease: A Historical Perspective. Lancet 27, epub (2013), doi:10.1016/S0140-6736(13)61752-3Google Scholar
- 6.Silverman, B.: Density Estimation for Statistics and Data Analysis. Chapman and Hall (1992)Google Scholar