“AI Surpasses Doctors in Eye Problem Assessment, Research Reveals”
New findings from the University of Cambridge have demonstrated that artificial intelligence (AI) might be more capable than doctors at diagnosing eye conditions. The study utilized GPT-4, the AI model behind the widely recognized chatbot, ChatGPT, and discovered it could outperform certain doctors in assessing eye problems.
The comparative research involved testing GPT-4 against medical professionals from various levels, including trainee and expert eye doctors, as well as unspecialised junior doctors. The AI significantly outscored the junior doctors, while its performance was comparable with that of specialised ophthalmologists.
While the study also shows that top-performing doctors still outpaced the AI, the study lead, Dr Arun Thirunavukarasu, emphasized AI's potential to aid in triage and benefit the healthcare sector when specialist access is limited.
Sharing his perspective, Dr Thirunavukarasu stated, “Even considering AI's future role, doctors will remain at the helm of patient care. The priority should always lie in empowering patients to decide if they want AI involvement.”
He further argued the need for “practicing doctors” to be considered as a benchmark for AI’s performance, as opposed to “sets of exam results”. This opinion underscores the creative approach of their research, ultimately seeking an authentic measure of AI's utility in real-world scenarios.
In favor of AI? Or not? Share your thoughts!
To learn more about AI transformations across various fields, check out our sponsor, Klap AI's Artificial Intelligence platform at www.TheBestAI.org/now .
#AIInHealthcare #FutureOfMedicine #ChatGPT #CambridgeStudy
