Artificial Intelligence Applications in Sickle Cell Retinopathy Imaging: Current Progress, Challenges, and Future...

Area of Science:

• Ophthalmology
• Medical Imaging
• Artificial Intelligence

Background:

• Sickle cell retinopathy (SCR) is a primary cause of vision loss in sickle cell disease patients.
• Current SCR detection and monitoring face challenges due to complex retinal changes and reliance on expert interpretation.
• Artificial intelligence (AI) has demonstrated expert-level performance in analyzing retinal images for various eye conditions.

Purpose of the Study:

• To review recent advancements in AI applications for sickle cell retinopathy imaging.
• To highlight the potential of AI in improving SCR diagnosis, staging, and monitoring.
• To identify future opportunities for the clinical translation of AI-based tools in SCR management.

Main Methods:

• Systematic literature review adhering to PRISMA guidelines.
• Comprehensive searches conducted in PubMed/MEDLINE, Embase, and Web of Science databases.
• Analysis of studies employing classical machine learning and deep learning algorithms for SCR detection and classification from ophthalmological images.

Main Results:

• Four studies met the inclusion criteria for the review.
• Two studies utilized classical machine learning for SCR classification based on extracted imaging features.
• Four studies employed deep learning algorithms for detecting SCR features in ophthalmological images, with one study differentiating SCR from other retinal diseases.

Conclusions:

• Deep learning holds significant potential for enhancing SCR detection, staging, and monitoring across various imaging modalities.
• Further research is essential to facilitate the clinical adoption of AI-driven SCR diagnostic tools.
• AI-based solutions can potentially improve diagnostic precision, personalize patient care, and enhance outcomes for individuals with sickle cell retinopathy.