Neural network and principal component analyses of highly variable myocardial mechanical waveforms derived from echocardiographic ultrasound images

We introduce a new type of data for classification of regional segments of myocardium. We have analyzed strain measurements taken throughout the cardiac cycle from the echocardiograms of pigs. Classifications by both principal component analysis (PCA) and by neural network (NN) are combined for a data mining operation. Differences in strain waveforms between normal and diseased myocardium may further elucidate the corresponding changes in physiology. Altered functioning of the heart muscle is reflected by strain, and objective computer analysis should aid in the diagnosis of ischemia. We hypothesize that the entire strain waveform over one heart cycle can be classified to functionally determine whether or not a myocardial region is perfused.

[1]  J. Seward,et al.  Both systolic and diastolic dysfunction characterize nonischemic inhibition of myocardial energy metabolism: an experimental strain rate echocardiographic study. , 2004, Journal of the American Society of Echocardiography.

[2]  J F Greenleaf,et al.  Real-time strain rate echocardiographic imaging: temporal and spatial analysis of postsystolic compression in acutely ischemic myocardium. , 2001, Journal of the American Society of Echocardiography : official publication of the American Society of Echocardiography.

[3]  T. Abraham,et al.  Strain and strain rate echocardiography , 2002, Current opinion in cardiology.

[4]  Armando Manduca,et al.  Image Analysis Using Modified Self-Organizing Maps: Automated Delineation of the Left Ventricular Cavity Boundary in Serial Echocardiograms , 1996, VBC.

[5]  A. Støylen,et al.  Real-time strain rate imaging of the left ventricle by ultrasound. , 1998, Journal of the American Society of Echocardiography : official publication of the American Society of Echocardiography.

[6]  P. Lysyansky,et al.  Global longitudinal strain: a novel index of left ventricular systolic function. , 2004, Journal of the American Society of Echocardiography : official publication of the American Society of Echocardiography.

[7]  James F. Greenleaf,et al.  Higher myocardial strain rates duringisovolumic relaxation phase than duringejection characterize acutely ischemic myocardium , 2002 .

[8]  J. Seward,et al.  Higher myocardial strain rates during isovolumic relaxation phase than during ejection characterize acutely ischemic myocardium. , 2002, Journal of the American College of Cardiology.

[9]  Marek Belohlavek,et al.  Time to onset of regional relaxation: feasibility, variability and utility of a novel index of regional myocardial function by strain rate imaging. , 2002, Journal of the American College of Cardiology.

[10]  Zvi Vered,et al.  Two-dimensional strain-a novel software for real-time quantitative echocardiographic assessment of myocardial function. , 2004, Journal of the American Society of Echocardiography : official publication of the American Society of Echocardiography.

[11]  Milan Sonka,et al.  Left ventricle contour detection in x-ray angiograms using multi-view active appearance models , 2003, SPIE Medical Imaging.

[12]  C. Goodall.  Procrustes methods in the statistical analysis of shape , 1991 .

[13]  Milan Sonka,et al.  Automatic segmentation of echocardiographic sequences by active appearance motion models , 2002, IEEE Transactions on Medical Imaging.

[14]  T. Abraham,et al.  Myocardial contractility by strain echocardiography: comparison with physiological measurements in an in vitro model. , 2003, American journal of physiology. Heart and circulatory physiology.

[15]  Milan Sonka,et al.  Multistage hybrid active appearance model matching: segmentation of left and right ventricles in cardiac MR images , 2001, IEEE Transactions on Medical Imaging.

[16]  I. Hashimoto,et al.  Myocardial strain rate is a superior method for evaluation of left ventricular subendocardial function compared with tissue Doppler imaging. , 2003, Journal of the American College of Cardiology.

[17]  Brian Young,et al.  Added value of new acute coronary syndrome computer algorithm for interpretation of prehospital electrocardiograms. , 2004, Journal of electrocardiology.

[18]  Milan Sonka,et al.  3-D active appearance models: segmentation of cardiac MR and ultrasound images , 2002, IEEE Transactions on Medical Imaging.

[19]  Brenda K. Wiederhold,et al.  ECG to identify individuals , 2005, Pattern Recognit..

[20]  Armando Manduca,et al.  Extraction of Endocardial Boundary From Echocardiographic Images by Means of the Kohonen Self-Organizing Map , 1996 .

[21]  Marek Belohlavek,et al.  Quantitation of regional myocardial function during short-lived events with ultrasound imaging , 2004, 2004 2nd IEEE International Symposium on Biomedical Imaging: Nano to Macro (IEEE Cat No. 04EX821).