10.46243/jst.2021.v6.i3.pp155-168 registered

As population aging has become a major demographic trend around the world, patients suffering from eye diseases, such as Glaucoma, ARMD are expected to increase. Early detection and appropriate treatment of eye diseases are of great significance to prevent vision loss and promote living quality. Conventional diagnosis methods are tremendously dependent on physicians, professional experience and knowledge, which lead to high misdiagnosis rate and huge waste of medical data. In this project, a deep learning model-based method which is inspired by the diagnostic process of human ophthalmologists is proposed to automatically classify the fundus photographs into 2 types with or without ARMD categories also, with or without Glaucoma. The project consists of two different neural network models developed to recognize the diseases, Glaucoma and ARMD.Better accuracy is obtained as we use deep learning. This project will be an aid to eye specialists in giving an efficient treatment. Eyesight is one of the most important senses, the developed project can help people all over to maintain eye care. This project uses Kaggle Glaucoma and ARMD datasets. This model predicts Glaucoma with 90% accuracy and ARMD with more than 70% accuracy.