The computers that can spot eye disease
94% success rate is as good as best doctors
COMPUTERS are now able to diagnose eye diseases as accurately as experts – potentially helping save the sight of millions.
British scientists have developed an artificial intelligence system that can spot eye diseases with 94 per cent accuracy.
For those who need urgent referrals, the technique promises to slash the waiting time to see a consultant from weeks to days.
Blindness can often be prevented by early detection and treatment. Researchers believe the scanning system could be in use in 30 British hospitals and clinics in less than three years.
The breakthrough is a collaboration between London’s Moorfields Eye Hospital NHS Foundation Trust, UK-based Google Deep Mind, UCL, and Southampton University. Dr Pearse Keane, of the Moorfields team, said: ‘The number of eye scans we’re performing is growing at a pace much faster than human experts are able to interpret them.
‘There is a risk that this may cause delays in the diagnosis and treatment of sightthreatening diseases, which can be devastating.’
Dr Keane added: ‘ The AI technology we’re developing is designed to prioritise patients who need to be seen and treated urgently by a doctor or eye care professional.’
Data from almost 15,000 anonymous NHS patients was used to help the AI system learn how to spot ten key features of eye disease from complex retinal scans. These scans use light rather than X-rays or ultrasound to generate 3D images of the back of the eye, revealing abnormalities that may point to disease.
Findings published in the journal Nature Medicine showed the AI system was able to triage more than 50 eye conditions correctly in more than 94 per cent of cases. The technology does not make a definitive diagnosis but, on the basis of signs such as blocked retinal veins, it can swiftly recommend which patients should be seen urgently by a specialist or simply placed under observation. Scientists are now planning clinical trials.
The technology could be introduced at all 30 Moor- fields hospitals in two and a half years. More than 285million people live with a form of sight loss, including more than two million in the UK.
Mustafa Suleyman, of Deep-Mind Health, said: ‘We believe artificial intelligence can help solve some of society’s biggest health challenges.
‘These incredibly exciting results take us one step closer to that goal and could, in time, transform the diagnosis, treatment and management of patients, not just at Moorfields, but around the world.’
Derek Hill of UCL said: ‘These papers provide further evidence that artificial intelligence will soon be routinely supporting doctors in streamlining diagnosis and treatment of numerous illnesses.’
But Dr Larisa Soldatova of Goldsmiths, University of London, said: ‘As Deep Mind is owned by Google, a company with a poor reputation for customer confidentiality, there are understandable deep concerns.’