Using Fundus Photography for Early Glaucoma Diagnosis by VGG-16 Architecture

Glaucoma stands as a prominent cause of irreversible vision impairment globally, marked by gradual harm to the optic nerve leading to visual field abnormalities. Timely identification and precise diagnosis play a pivotal role in ensuring effective treatment and the preservation of eyesight. The classification of the medical history of patients affected by glaucoma involves the utilization of a combination of machine learning algorithms, namely RandomForest and XGBoost, to identify individuals at risk. This research article aims to enhance the diagnostic capabilities of Glaucoma through the optimized utilization of contemporary methodologies, with a focus on leveraging deep learning techniques, specifically the VGG-16 model, in the analysis of Fundus Photograph captured by a panoptic direct ophthalmoscope. The novelty of this research lies in its integration of an individual's medical history, coupled with the detailed analysis of fundus photography, to enhance diagnostic precision and medical insights. In conclusion, the research is integrated with a mobile application for timely and accurate diagnosis of glaucoma which is essential in preventing further vision loss.