Neural Networks Unravel the Early Warnings of Eye Disease

The most common cause of blindness in the industrialized nations is glaucoma. Unfortunately, a reliable diagnosis in the early stages of the disease cannot be achieved. Researchers of the EADQ cooperate with the university clinic, eye hospital, in Erlangen to promote early detection of glaucoma. This project was sponsored by Deutsche Forschungsgemeinschaft (DFG) till March 1997. It uses a commercial laser scanning tomograph to investigate the fundus of the eye. This devices measures the morphology in terms of quantiative features. The neural approach by EADQ classifies the feature set in terms of principal glaucoma stages. Thus, it opens the way to more quantitative evaluation using a finer classification scheme