DocumentCode :
2754050
Title :
Neural network and principal component analyses of highly variable myocardial mechanical waveforms derived from echocardiographic ultrasound images
Author :
McMahon, Eileen M. ; Korinek, Josef ; Zhang, Honghai ; Sonka, Milan ; Manduca, Armando ; Belohlavek, M.
Author_Institution :
Dept. of Internal Medicine, Mayo Clinic Coll. of Medicine, Rochester, MN, USA
Volume :
5
fYear :
2005
fDate :
31 July-4 Aug. 2005
Firstpage :
3017
Abstract :
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.
Keywords :
data mining; echocardiography; image classification; medical image processing; muscle; neural nets; patient diagnosis; principal component analysis; waveform analysis; data mining; diseased myocardium; echocardiographic ultrasound image; heart muscle; myocardial mechanical waveform; neural network; objective computer analysis; principal component analysis; strain waveform; Capacitive sensors; Data mining; Heart; Image segmentation; Myocardium; Neural networks; Physiology; Principal component analysis; Strain measurement; Ultrasonic imaging;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks, 2005. IJCNN '05. Proceedings. 2005 IEEE International Joint Conference on
Print_ISBN :
0-7803-9048-2
Type :
conf
DOI :
10.1109/IJCNN.2005.1556406
Filename :
1556406
Link To Document :
بازگشت