Fusing Age-Guided Features in Convolutional Neural Network for Accurate Retinal Disease Diagnosis

Eye diseases are a significant global health issue, impacting millions of people and potentially leading to serious side effects, such as vision loss, in the absence of proper treatment and early diagnosis. Nevertheless, the traditional method faces several challenges, including time-consuming procedures, susceptibility to human error, and workforce shortage in impoverished areas. There is a growing demand for accurate automation systems to address these issues. To address the aforementioned problem, extensive research has been conducted, leveraging machine learning techniques to develop an eye disease recognition system. Previous studies demonstrated the potential of convolutional neural networks in diagnosing the patient. However, these methods often suffer from inaccuracies and limitations in targeting a wide range of diseases, rendering them less practical. Thus, I propose a novel eye disease recognition framework to overcome the previously mentioned shortage. I introduce an age-guided approach to improve the accuracy of the disease diagnosis process. The research results demonstrate the outperformance of the proposed model, achieving state-of-the-art performance with an accuracy of 84% on a publicly available eye disease dataset.