Implementation of Artificial Intelligence in Retinopathy of Prematurity Care: Challenges and Opportunities

Area of Science:

• Ophthalmology
• Medical Imaging
• Artificial Intelligence

Background:

• Retinopathy of prematurity (ROP) diagnosis relies heavily on medical imaging.
• The rising incidence of ROP, particularly in low and middle-income countries (LMICs), strains healthcare resources.
• Existing healthcare systems face significant challenges in implementing AI for ROP screening.

Purpose of the Study:

• To review current AI and imaging systems for ROP diagnosis.
• To highlight the importance of telemedicine infrastructure for AI implementation in ROP screening.
• To discuss successful ROP program models in various economic settings and identify future research needs.

Main Methods:

• Literature review of available AI and imaging technologies for ROP.
• Analysis of implementation barriers including infrastructure, regulatory, legal, cost, sustainability, and scalability.
• Examination of successful ROP screening program case studies from high-income countries (HICs) and LMICs.

Main Results:

• AI systems show promise for ROP diagnosis, but require validation across diverse populations and with low-cost imaging devices.
• A stable telemedicine infrastructure is essential for the successful deployment of AI-driven ROP screening.
• Effective ROP screening programs can be implemented in both HICs and LMICs, demonstrating scalability and adaptability.

Conclusions:

• Further research is needed to validate AI systems using varied populations and cost-effective imaging technologies.
• Sustainable and affordable ROP screening programs are critical for preventing childhood blindness globally.
• Integrating AI and robust telemedicine infrastructure offers a scalable solution to improve ROP detection and management.