Eye disease is one of the most common causes of sight loss, but in about 80% of cases, this can be prevented by early detection and treatment.
A novel artificial intelligence (AI) system can diagnose eye disease as accurately as expert ophthalmologists. The technology can recommend the correct referral decision for more than 50 types of eye disease with 94% accuracy.
Dr Pearse Keane, a consultant ophthalmologist at Moorfields Eye Hospital and NIHR clinician scientist at the UCL Institute of Ophthalmology, collaborated with Google-owned company DeepMind to develop the pioneering AI system. They created an algorithm to analyse optical coherence tomography (OCT), a high-resolution 3D scan of the retina. Using almost 15,000 historic and anonymous OCT scans, they taught the computer how to spot signs of eye disease, including haemorrhages, lesions and fluid build-up.
The system could be of enormous benefit to the NHS and has the potential to transform the diagnosis and treatment of eye disease patients across the world. The technology could also be applied to x-ray and MRI scans to detect cancer and other diseases.
Related links:
Breakthrough in AI technology to improve care for patients
Artificial technology ‘did not miss a single urgent case’
Clinically applicable deep learning for diagnosis and referral in retinal disease
