Begin typing your search...

    AI-powered models match ophthalmologists in diagnosing eye infections: Study

    AI-powered models match ophthalmologists in diagnosing eye infections: Study
    X

    Representative Image 

    NEW DELHI: Artificial Intelligence (AI)-powered models are at par with eye care specialists in diagnosing infectious keratitis (IK) -- a leading cause of corneal blindness worldwide, finds a study on Tuesday.

    The study further established the potential of AI and deep learning models in boosting healthcare.

    Infectious keratitis, commonly known as corneal infection, resulted in about 5 million cases of blindness worldwide and accounted for about 2 million cases of monocular blindness annually. It has particularly affected people in low- and middle-income countries.

    Researchers from the University of Birmingham in the UK conducted a meta-analysis of 35 studies that utilised Deep Learning (DL) models to diagnose infectious keratitis.

    Their findings, published in eClinicalMedicine, showed that AI models matched the diagnostic accuracy of ophthalmologists. Compared to ophthalmologists' 82.2 per cent sensitivity and 89.6 per cent specificity, the AI models exhibited a sensitivity of 89.2 per cent and specificity of 93.2 per cent.

    “Our study shows that AI has the potential to provide fast, reliable diagnoses, which could revolutionise how we manage corneal infections globally,” Dr. Darren Ting, Consultant Ophthalmologist at the University of Birmingham.

    Ting noted that the AI-powered models may be of significant benefit in areas “where access to specialist eye care is limited and can help to reduce the burden of preventable blindness worldwide”.

    The AI models also proved effective at differentiating between healthy eyes, infected corneas, and the various underlying causes of IK, such as bacterial or fungal infections.

    However, the researchers called for interpreting the results with caution due to the image-based analysis that did not account for potential correlation within individuals.

    They emphasised the need for more diverse data and further external validation to increase the reliability of these models for clinical use.

    --IANS

    rvt/

    IANS
    Next Story