Artificial intelligence for improving sickle cell retinopathy diagnosis and management

Area of Science:

• Ophthalmology
• Medical Imaging
• Artificial Intelligence

Background:

• Sickle cell retinopathy (SCR) often progresses asymptomatically, leading to vision loss from complications like vitreous hemorrhages and retinal detachments.
• Current screening methods face challenges in access and adherence, particularly in underserved areas disproportionately affected by sickle cell disease.
• There is a critical need for innovative screening approaches to detect vision-threatening SCR.

Purpose of the Study:

• To review current literature on using artificial intelligence (AI) with multimodal retinal imaging for automated SCR screening.
• To explore future research directions for AI-driven, imaging-based SCR screening.
• To discuss the development of machine learning models for quantitative tracking of SCR progression.

Main Methods:

• Literature review of existing studies on AI and multimodal retinal imaging in SCR screening.
• Analysis of recent advancements in machine learning for disease progression monitoring.
• Discussion of potential AI applications for clinical diagnosis and management.

Main Results:

• AI coupled with multimodal retinal imaging shows potential for automated, accurate screening of SCR.
• Machine learning models are being developed to quantitatively track SCR progression over time.
• AI applications can inform evidence-based and resource-efficient clinical diagnosis and management.

Conclusions:

• AI-powered, imaging-based screening can significantly expand access to early detection of SCR.
• Machine learning offers tools for objective monitoring of SCR, addressing variability in specialist practices.
• These AI applications hold great promise for improving the clinical diagnosis and management of SCR.