How Does AI Compare to Ophthalmologists in Diagnosing Infectious Keratitis?

Artificial intelligence (AI) may soon play a role in diagnosing infectious keratitis (IK), a leading cause of corneal blindness, according to a recent meta-analysis published in eClinicalMedicine. Researchers found that deep learning (DL) models matched or exceeded the diagnostic accuracy of ophthalmologists, offering a promising solution for regions with limited access to specialized eye care.

Key Findings on AI Diagnostic Accuracy

In the meta-analysis led by Dr. Darren Ting of the University of Birmingham, researchers reviewed 35 studies involving DL models for diagnosing IK. These AI models, trained on over 136,000 corneal images, demonstrated a sensitivity of 89.2% and specificity of 93.2%, compared to ophthalmologists' sensitivity of 82.2% and specificity of 89.6%.

The study also revealed that DL models excelled at distinguishing between healthy corneas, infectious keratitis cases, and underlying causes of infection, such as bacterial or fungal pathogens. These findings suggest AI could support ophthalmologists in rapidly and reliably identifying corneal infections in clinical settings.

"AI has the potential to provide fast, reliable diagnoses, which could revolutionize how we manage corneal infections globally," stated Dr. Ting. He highlighted the significant implications for low- and middle-income countries, where access to specialized eye care remains limited.

Addressing Global Corneal Blindness

Infectious keratitis affects millions worldwide and is a leading cause of preventable blindness, particularly in underserved regions. Early and accurate diagnosis is critical to preventing vision loss. By offering diagnostic accuracy comparable to specialists, AI could bridge gaps in care, reducing the burden of IK-related blindness.

However, the study's authors emphasized the importance of further validation using diverse datasets to ensure the reliability of AI models in varied clinical settings. As AI technology advances, its integration into eye care could become a game-changer in preventing corneal blindness on a global scale.