Robust Extraction of the Optic Nerve Head in Optical Coherence Tomography

Abstract

Glaucoma is a leading cause of blindness. While glaucoma is a treatable and controllable disease, there is still no cure available. Early diagnosis is important in order to prevent severe vision loss. Many current diagnostic techniques are subjective and variable. This provides motivation for a more objective and repeatable method. Optical Coherence Tomography (OCT) is a relatively new imaging technique that is proving useful in diagnosing, monitoring, and studying glaucoma. OCT, like ultrasound, suffers from signal dependent noise which can make accurate, automatic segmentation of images difficult. In this article we propose a method to automatically extract the optic nerve and retinal boundaries from axial OCT scans through the optic nerve head. We also propose a method to automatically segment the curve to extract the nerve head profile that is important in diagnosing and monitoring glaucoma.