AUTOMATED DIAGNOSIS OF GLAUCOMA

Abstract

This Thesis applies the process and knowledge of digital signal processing and image processing to diagnose glaucoma using digital fundus images.

The algorithm was tested with a separate set of 60 fundus images. Digital fundus image has been widely used for automatic detection of the optic disc, blood vessels and the computation of the features. Features such as the blood vessels ratio, optic disc and the ratio of the blood vessel area in the inferior-superior side to area of blood vessel in the nasal-temporal side were extracted. A Gaussian mixture model classifier was used as the classification tool.

In addition to diagnose of glaucoma, a graphical user interface (GUI) was developed during this work. The GUI is for automatic diagnosing and displaying the diagnosis result in a friendlier user interface.