Paper Info
Title
Study on Ultrasound Kidney Images Using Principal Component Analysis: A Preliminary Result
Abstract
An effort has been made to test the potential of principal component analysis (PCA) method for quantifying and classifying the ultrasound kidney images. For our analysis two different classes of kidney images namely normal (NR) and medical renal diseases (MRD) are considered. The eigen values and vectors are derived for a set of 40 images. The weight vectors (WV) are estimated from the obtained eigen vectors. The result indicates that the mean value of WV for NR is 0.4205 and for MRD is -0.9983. The sum value of WV for NR and MRD is 8.4103 and -19.9657 respectively. The regression analysis shows that WV's of two classes are negative correlated and are in weak moderate association (-0.4271). The student t-test specifies that the eigen value and WV are much significant (p<0.005) in separating the classes of kidney image under study. These analysis shows that there exists an appreciable measure of relevance for this parameter weight vector in classifying the kidney images. .
Year
Venue
Keywords
2004
ICVGIP
regression analysis,principal component analysis,ultrasound
Field
DocType
Citations 
Mean value,Pattern recognition,Regression analysis,Computer science,Weight,Artificial intelligence,Eigenvalues and eigenvectors,Principal component analysis,Ultrasound
Conference
5
PageRank 
References 
Authors
0.87
8
3
Name
Order
Citations
PageRank
C. Karthikeyini150.87
K. Bommanna Raja2374.33
M. Madheswaran310215.57