An artificially intelligent (AI) program that is "trained" with data from thousands of NHS patients can identify important signs of eye disease and the world's top experts, a study has shown.
The system has the potential to prevent irreversible vision loss by ensuring early treatment of patients with the most severe eye diseases
Initial results of scanning technology tests are so promising that they could be implemented in 30 UK hospitals within three years  Dr. med. Pearse Keane analyzes an OCT retinal scan (Moorfields Eye Hospital / PA)
Dr. Pearse Keane, a senior member of the team at the Moorfields Eye Hospital NHS Foundation Trust in London, said: "The number of eye scans we perform is growing much faster than human experts can interpret.
" There is a risk that This will lead to delays in the diagnosis and treatment of vision-impairing diseases that "The AI technology we develop aims to prioritize patients who need to be seen and treated urgently by a doctor or ophthalmologist.
"If we can diagnose and treat eye diseases early, it gives us the best chance of preserving people's eyesight.
Anonymous diagnostic data from nearly 15,000 NHS patients were used to help AI learn "brain" how to treat 10 Key features of eye disease from complex optical coherence tomography (OCT) to detect retinal scans.
An OCT scan uses light instead of X-rays or ultrasound to generate 3D images of the back of the eye, abnormalities that are signs of disease
The new system was developed by scientists at Moorfields, Un The results published in the journal Nature Medicine showed that in more than 94% of cases, patients with more than 50 eye diseases were able to correctly allocate what the Leis It was matched by leading experts from around the world.
The program does not generate a definitive diagnosis. But due to clinical symptoms such as holes in the macula, the central retinal region, or blocked retinal veins, it can be quickly recommended which patients should be urgently visited by a specialist or simply observed.
Under a In a five-year development program, the scientists now want to advance clinical trials.
Possibly, the system could be implemented in all 30 UK hospitals and clinics in Moorfields in just two and a half years – this is a very rough estimate and not a fixed target.
Over 285 million people worldwide are living with some type of vision loss, including more than two million people in the UK.
In many cases, blindness can be prevented by the early detection and treatment of eye diseases such as macular degeneration and glaucoma.
Mustafa Suleyman, co-founder and head of Applied AI at DeepMind Health, said, "We use DeepMind Health because we believe that artificial intelligence can help solve some of society's biggest health challenges, such as preventable vision loss Affecting millions of people worldwide
"These unbelievably exciting results bring us one step closer to this goal Over time, the diagnosis, treatment and treatment of patients with ocular eye diseases, not only in Moorfields but around the world, be transformed.
The study was one of two that attested the effectiveness of AI screenings of patients in Nature Medicine
The other program, which was developed in the US, was able to detect signs of brain damage from stroke, hemorrhage and hydrocephalus by computed tomography (CT) in just 1.2 seconds.
Lead researcher Dr. Eric Oermann of the Mount Sinai Hospital in New York City said, "Such a triage system can alert physicians to a critical finding, which otherwise remains in minutes in a queue s.
Derek Hill, a professor of medical imaging at the University Hospital, said, "These papers provide further evidence that artificial intelligence will soon routinely assist physicians in optimizing the diagnosis and treatment of many diseases. "