Artificial intelligence for retinal diseases

Area of Science:

• Ophthalmology
• Medical Artificial Intelligence
• Biomedical Imaging

Background:

• Common retinal diseases pose a significant threat to vision globally.
• Early diagnosis and effective management are crucial for preventing vision loss.

Purpose of the Study:

• To review worldwide applications of artificial intelligence (AI) in diagnosing and managing common retinal diseases.
• To explore AI's potential impact on treatment outcome analysis and disease management.

Main Methods:

• Conducted an online literature review of AI applications for retinal diseases using PubMed Central.
• Included studies on screening, diagnosis, monitoring, and treatment outcomes for conditions like AMD, DR, ROP, and SCR.
• Focused on AI's use with retinal imaging techniques such as OCT and OCTA.

Main Results:

• AI demonstrates utility in screening for DR, AMD, ROP, and SCR.
• AI algorithms can predict disease progression and treatment response using validated datasets.
• AI facilitates rapid, quantitative analysis of retinal biomarkers from OCT and OCTA, and may assist in robotic surgery planning.

Conclusions:

• AI tools can assist clinicians in disease screening, monitoring, and quantitative analysis of treatment outcomes.
• AI may help reduce socioeconomic disparities affecting retinal disease outcomes.
• The public health impact of AI in preventing blindness from retinal diseases requires further determination.