Machine Learning for Early Detection of Eye Diseases

Fundus images enable diagnosing eye conditions such as diabetic retinopathy, macular degeneration, glaucoma etc. However, challenges arise when patients have multiple medical conditions that present conflicting symptoms, complicating the diagnostic process. Additionally, early-stage symptoms may be too vague to differentiate between diseases, making early detection difficult. This work focuses on the usage of statistics derived from fundus images to enable multi-disease classification.