Role of Artificial Intelligence in Detecting and Grading Cataracts Using Color Fundus Photographs: a Systematic Review and Meta-analysis

Abstract

Background      : Cataracts are a leading cause of blindness and visual impairment worldwide, affecting millions of people. Early detection and accurate grading of cataracts are critical for timely intervention and improving patient outcomes. Artificial intelligence (AI), particularly deep learning, has emerged as a powerful tool for automating the detection and grading of cataracts using color fundus photographs.

Methods           : A systematic review and meta-analysis was undertaken in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines; a thorough literature search through databases such as PubMed, IEEE Xplore, and Google Scholar was conducted. The search parameters were restricted to studies published within the time frame of January 2020 to March 2025.

Results             : A total of six studies were included in this systematic review and meta-analysis. Utilizing DTA meta-analysis, sensitivity ranged from 0.88 to 0.99, while specificity ranged from 0.89 to 0.99. Diagnostic Odds Ratio was estimated at 88.5, indicating that patients with cataracts are nearly 89 times more likely to be correctly identified by the AI model than non-cataract patients being misclassified.

Conclusion        : AI particularly deep learning, has made significant strides in detecting and grading cataracts using color fundus photographs. The high accuracy, cost-effectiveness, and accessibility of AI models make them a valuable tool for improving cataract screening and management. As research continues to advance, AI has the potential to revolutionize cataract care, enabling early detection and timely intervention for millions of people worldwide.