UUM Electronic Theses and Dissertation
UUM ETD | Universiti Utara Malaysian Electronic Theses and Dissertation
FAQs | Feedback | Search Tips | Sitemap

Integrated Retinal Information System for Analyzing Kidney Condition

Perdana, Hatta (2009) Integrated Retinal Information System for Analyzing Kidney Condition. Masters thesis, Universiti Utara Malaysia.

[thumbnail of Hatta_Perdana_2009.pdf] PDF
Restricted to Registered users only

Download (1MB) | Request a copy
[thumbnail of 1.Hatta_Perdana_2009.pdf]

Download (114kB) | Preview


Iridology is a science and practice that can express body state based on the analysis of iris structure. The changes or disturbances of disease on body network will be informed by neuron nerve fiber to brain. This energy wave information spread to eye by brain, recorded and fixed by pupil.Then, these recorded fixation become data trails which can be detected by disturbance/disease that is filed by body organ. The research about iridology to analyzing kidney condition has been conducted before using Learning Vector Quantization (LVQ) method. The accuracy is not 100%.
In this research, the researcher implements Support Vector Machine(SVM) in classifying the kidney condition to replace LVQ using Matlab R2007b. The accuracy in classifying the kidney condition for right eyes is 100% and for the left eyes is 100% in training set data. If we compared to
the accuracy of classification using LVQ, implementing SVM is much better because by implementing LVQ, the accuracy is only 96% for right eyes and only 92% for left eyes.

Item Type: Thesis (Masters)
Supervisor : UNSPECIFIED
Item ID: 1571
Uncontrolled Keywords: Data Mining, System Development, Integrated Retinal Information System, Iris Diagnosis, Kidney Condition
Subjects: Q Science > QA Mathematics > QA76 Computer software > QA76.76 Fuzzy System.
Divisions: College of Arts and Sciences (CAS)
Date Deposited: 23 Feb 2010 04:03
Last Modified: 24 Jul 2013 12:12
Department: College of Art and Sciences
URI: https://etd.uum.edu.my/id/eprint/1571

Actions (login required)

View Item
View Item