Effect of eye diseases on the retina and iris pattern recognition and its significance in early ophthalmic diagnosis

The paper presents a new approach to retinal and iris pattern recognition for human identification/ identity verification and early detection of ocular diseases. The imaging method is based on biometric properties of iris and retina. The innovative element involves the assessment of effect of eye diseases on iris and retinal pattern. The research was carried out in close cooperation with IT specialists. An algorithm was developed for retinal and iris pattern recognition in patients with various eye diseases in an attempt to determine whether biometric scanning is capable of identifying affected patients and which conditions preclude accurate identification. It was shown that in advanced stages, retinal vascular diseases (retinal detachment, retinal vein and artery occlusion) made it impossible to identify an individual based on recognising their retinal pattern. This effect was fully or partly reversible following treatment. Iris pattern was disturbed by extensive and deforming lesions, such as tumors and hypopyon. It was proved that hyphema, previous glaucoma surgery or cataract did not interfere with iris pattern recognition.