A breakthrough research from Moorfields Eye Hospital NHS Foundation Trust, DeepMind Health and the University College London, reports about an artificial intelligence system that could spot signs of eye disease as accurately as expert doctors and help recommend how patients should be referred for care. These findings were published in the journal Nature Medicine.
Researchers used optical coherence tomography scans from referred patients. Machine learning technology was used to train the artificial intelligence system to identify 10 features of more than 50 eye diseases from these scans. Based on the most urgent conditions detected, the system recommended a referral decision. Clinicians also viewed the same scans and made their own referral decisions which established whether the artificial intelligence system was making correct referrals.
Artificial intelligence was able to make the right referral recommendation more than 94% of the time. The system made no clinically serious wrong decisions and performed comparably to the 2 best performing retina specialists. Importantly, the system could provide information that could explain eye care professionals how it arrived at its recommendations. Also, it could be easily applied to different types of eye scanner, thus significantly increasing number of people who could benefit from it.
Results from clinical trials would further explain how this technology improves patient care in practice. Mustafa Suleyman, Co-founder and Head of Applied Artificial Intelligence at DeepMind Health commented: “These incredibly exciting results take us one step closer in addressing unavoidable sight loss and could, in time, transform the diagnosis, treatment and management of patients with sight threatening eye conditions around the world.”
