Comparison of Classification Effects between Principal Component and Multilinear Principal Component Analysis for Study of Cardiology Ultrasound in Left Ventricle
| Autor: | Wei-Yi Liu, 劉瑋毅 |
|---|---|
| Rok vydání: | 2013 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 101 In this work, we study heart diseases related to diastolic and systolic of heart in left ventricle. Following Chen(2012) and Yang(2012), the differences between the gray-scale values of diastolic and systolic in left ventricle are uesd to evaluate the function of heart. As the ultrasound image data for each subject is expressed as a highdimensional matrix, the usual principal component analysis and the multilinear principal component analysis by Hung et al(2012) are used for dimension reduction and the corresponding loadings in each factor component are compared to see the major areas in the image with larger differences. Later the factor scores for the subject are used to carry out the regularized discriminant analysis given in Friedman(1989). The Local Search Procedure to select the appropriate variables similar as in Pacheco et al.(2006) is also used. According to the above analysis methods, we have identified important areas in the heart for classification between normal and abnormal groups. The procedure has been applied to the heart ultrasound images and seems to have good performances in classification. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|