Anatomical digital image analysis of the angle and optic nerve – a novel method for glaucoma imaging
Abstract
Purpose
To devise a simple, fast and low-cost method for glaucoma imaging using anatomical digital image analysis of the angle and optic nerve in human subjects.
Methods
A pilot study included 10 open-angle glaucoma patients attending the glaucoma services at the Department of Ophthalmology, University of Szeged, Hungary; colour photographs of the fundus, standard optic nerve images with OCT and additional digital slit lamp images of the angle where collected. Digital image conversion and analysis of the angle using Image J (NIH, USA) and adaptive histogram equalization Contrast Limited AHE (CLAHE) to prevent noise amplification. Angle and optic nerve images, were analyzed separately in the Red, Green and Blue (RGB) channels, then 3D volumetric analysis of the degrees of angle, cup volume of the optic nerve took place. Horizontal tomogram reconstitution and nerve fiber detection methods were developed and compared to a standard Topcon 3D OCT.
Results
Angle measurements from digital angle images by gonioscopy all fitted within one standard deviation of Anterior OCT measurements. Comparative analysis of RGB channels produced volumetric cup and Horizontal tomogram results closely resembling those of OCT 3D appearance and B-scan of the cup, respectively. Similar RGB channel splitting and image subtraction produced a map closely resembling that of the thickness RNFL map on OCT.
Conclusions
While spectral domain OCT is rapidly progressing in the area of optic disc and chamber angle assessment, rising health care costs and lack of availability of the technology, opens demand for alternative forms of image analysis of glaucoma. Volumetric, geometric and segmentational data obtained through digital image analysis correspond well to those obtained by high definition OCT imaging. A larger number of subjects is needed to further validate the method.
