Google's DeepMind AI will identify ocular ailments at Moorfield Eye Hospital

Partnership will create a machine learning system to spot early occurrences of eye problems

Google's DeepMind artificial intelligence (AI) division has partnered with Moorfield Eye Hospital in London to create a machine learning system that can identify conditions that may threaten eyesight.

Moorfields will use the system to run algorithms on top of one million anonymous digital eye scans as a way to spot the early signs of conditions that may be missed by medical experts.

The DeepMind technology could also spot hidden problems that could result in harm to a patient's sight during long waiting times for specialist treatment.

"This is where DeepMind is able to help us understand these huge datasets and then put it together so that it benefits towards making a good diagnosis and achieving the best possible treatment for our patients," said Professor Sir Peng Tee Khaw, head of Moorfields' ophthalmology research centre.

The data being accessed by DeepMind will be kept anonymous to preserve patient privacy but still allow the machine learning technology to carry out work that will benefit the hospital.

The partnership with Moorfields is not DeepMind's first foray into the medical sector. Google's AI arm is also involved in the analysis of 1.6 million patent records held by Royal Free NHS Trust, despite there being some controversy over the company's access to such sensitive data.

DeepMind has previously gained acclaim after being used to create AlphaGo, an AI system that can beat champion human players of the highly complex Go board game that has often been described as something that machine learning systems cannot figure out.

Yet as AI technology gathers pace and becomes more advanced and easier to deploy thanks to the cloud, it is appearing in a wide range of diverse use cases, from unmanned coffee shops in London to 3D-printed driverless cars that can hold conversations with passengers.