AI may detect eye disease better than doctors, study finds

Credit: Unsplash+

In a recent study by the New York Eye and Ear Infirmary of Mount Sinai (NYEE), an advanced artificial intelligence (AI) system, specifically the latest generation called GPT-4 developed by OpenAI, has shown remarkable capabilities in the field of ophthalmology.

This study, published in JAMA Ophthalmology, highlights the AI’s potential to match or even surpass human ophthalmologists in diagnosing and treating glaucoma and retina diseases, conditions that affect millions worldwide.

The study involved a head-to-head comparison between the AI system and human eye specialists, including 12 attending specialists and three senior trainees from the Icahn School of Medicine at Mount Sinai.

They focused on a set of 20 questions from the American Academy of Ophthalmology—10 each for glaucoma and retina diseases—along with 20 deidentified patient cases from Mount Sinai-affiliated eye clinics.

The answers provided by both the AI and the human specialists were then analyzed for accuracy and completeness.

The findings were nothing short of impressive. The AI demonstrated a high level of proficiency, matching or exceeding the human experts’ performance.

It showed superior skills particularly in addressing glaucoma-related inquiries and case management, while also performing exceptionally well with retina questions, matching the accuracy of the human specialists and even surpassing them in providing comprehensive responses.

This study underscores the growing significance of AI in medicine, offering a glimpse into a future where AI tools could become indispensable allies for medical professionals.

By harnessing vast amounts of data, text, and images, AI systems like GPT-4 can support decision-making processes, potentially revolutionizing how diseases are diagnosed and treated.

In ophthalmology, where the volume of patients and the complexity of cases are high, AI’s impact could be particularly profound, enabling specialists to focus more on practicing evidence-based medicine.

Dr. Andy Huang, an ophthalmology resident at NYEE and the study’s lead author, shared his excitement about the AI’s performance, noting its potential to serve as a reliable assistant for eye specialists.

This could not only provide diagnostic support but also help alleviate the workload in managing complex cases or in situations where patient volume is high.

The study’s senior author, Dr. Louis R. Pasquale, highlighted the AI’s ability to enhance clinical documentation, drawing an analogy with how the AI application Grammarly improves writing skills.

Similarly, GPT-4 could offer valuable insights into improving clinical practices, particularly in documenting patient exam findings.

While further testing is necessary, the findings from this study point to a promising future for AI in ophthalmology.

For patients, the integration of AI into routine ophthalmic practice could mean quicker access to expert advice and more informed decision-making in their treatment plans, marking a significant step forward in the pursuit of advanced, AI-supported healthcare.

If you care about eye health, please read studies about how vitamin B may help fight vision loss, and MIND diet may reduce risk of vision loss disease.

For more information about eye disease, please see recent studies about how to protect your eyes from glaucoma, and results showing this eye surgery may reduce dementia risk.

The research findings can be found in JAMA Ophthalmology.

Copyright © 2024 Knowridge Science Report. All rights reserved.