Automatic Screening Algorithm for Narrow Anterior Chamber Angle and Angle-Closure Glaucoma Based on Slit-Lamp Image Analysis

Abstract

To perform a narrow anterior chamber angle (NACA) and angle-closure glaucoma (ACG) screening using the Van Herick’s method, the width of the peripheral anterior chamber depth (PACD) and corneal thickness have to be measured. These are the key parameters to identify a patient who may be suffering from ACG. Therefore, to develop an automatic screening algorithm for NACA and ACG based on slit-lamp image analysis, a width measurement algorithm has to be as accurate as possible in order to produce the best result. An algorithm was proposed to increase the screening accuracy and also to reduce the computational time. To overcome bright spot reflection on the cornea, the extraction of light reflected and the measurements of both the PACD and corneal thickness were improved. In order to remove unnecessary areas from the image and decrease the processing time, an improved algorithm was designed to automatically detect a region of interest (ROI). Then, only the ROI image was used in subsequent processes. Experimental results showed that the proposed algorithm is accurate and reliable.