Universiti Teknologi Malaysia Institutional Repository

Analisis statistik bentuk untuk pengkelasan data kraniofasial

Md. Hussin, Mohd. Bakery (2010) Analisis statistik bentuk untuk pengkelasan data kraniofasial. Masters thesis, Universiti Teknologi Malaysia, Faculty of Geoinformation and Real Estate.



Statistical shape analysis can be applied in many disciplines including medical field. In this study, the statistical shape analysis is employed for classification of craniofacial data. The craniofacial data of students are obtained using the Konica Minolta Vivid 910 laser scanner to produce the three dimensional craniofacial image. The three dimensional coordinates of the surface for each landmark are obtained digitally using RapidForm 2004 software. The methods used are Principal Component Analysis (PCA) and Generalized Procrustes Analysis (GPA), to obtain average configuration of facial point transformation and facial point average, respectively. The craniofacial data are divided into sex and age categories by using the statistical classification method. The craniofacial data are classified by three methods: (i) Procrustes examination analysis of 35 students with 18 landmarks each; (ii) Procrustes examination analysis of 76 data sets with 27 landmarks each, and; (iii) Examination analysis of the 35 data sets with 20 distance measurements each. The results showed that the Procrustes values of male data are larger than the female data, and the Procrustes values of those born in 1989 were larger than those born in 1991. Centroid size differences between the sex and age are significant, hence the values of PCA and GPA can be used for classification analysis. The research results produced the average value for craniofacial data based on age and sex.

Item Type:Thesis (Masters)
Additional Information:Supervisor : Prof. Dr. Halim Setan; Thesis (Sarjana Sains (Geoinformatik)) - Universiti Teknologi Malaysia 2010
Uncontrolled Keywords:face analysis, statistical shape
Subjects:R Medicine > RZ Other systems of medicine
Divisions:Geoinformation Science And Engineering (Formerly known)
ID Code:16273
Deposited By: Zalinda Shuratman
Deposited On:01 Feb 2012 07:31
Last Modified:13 Sep 2017 01:30

Repository Staff Only: item control page