The Role of Artificial Intelligence in the Diagnosis of Retinal Diseases

This is the authors' abstract. We don't add key points for this paper.

Background: Retinal diseases such as diabetic retinopathy, age-related macular degeneration, and glaucoma are major causes of vision loss and blindness around the world. Detecting these conditions early and treating them promptly is key to preventing them from worsening and saving sight. New developments in artificial intelligence (AI), especially in machine learning (ML) and deep learning (DL), now offer better ways to automatically analyze retinal images and help doctors make more accurate diagnoses in eye care. Methods: A narrative literature review was carried out using the PubMed database to find peer-reviewed studies on artificial intelligence in diagnosing retinal diseases. The review included recent studies that assessed AI algorithms used with retinal imaging methods like fundus photography and optical coherence tomography (OCT). Results: Multiple studies indicate that AI algorithms can accurately detect retinal diseases. Deep learning models, particularly convolutional neural networks, are effective at identifying features linked to diabetic retinopathy, age-related macular degeneration and glaucoma. Some AI systems for diabetic retinopathy screening have achieved sensitivities and specificities above 90%, which is comparable to experienced ophthalmologists. Furthermore, AI-based analysis of OCT images has shown promise in detecting structural retinal changes and monitoring disease progression. Conclusions: Artificial intelligence could greatly improve how retinal diseases are diagnosed and screened by allowing fast and accurate analysis of retinal images. Although there are still challenges with data quality, transparency, and clinical use, AI-based diagnostic systems will likely become an important part of future eye care.