Pearse Keane

What I’ve tried to do in my career is identify cool things in the tech world and see if they have applications in healthcare. 

In the NHS, since about 2017, ophthalmology is the busiest of all the medical specialties in terms of clinic appointments. There are 10 million appointments for eyes every year – that’s  ten percent of all appointments. Because we have so many patients to deal with, there are some patients that can lose sight and go blind because they can’t be seen or treated quickly enough.

My idea was that I would develop and apply the latest advances in artificial intelligence to identify eyes with the most sight-threatening disease, so we could prioritise those patients and get them in front of a doctor as soon as possible. I happened to read an article about the AI company DeepMind and that was the eureka moment. 

All Images © Moorfields Eye Hospital

Through our collaboration with DeepMind, we have developed an algorithm to assess retinal (OCT) scans for more than 50 diseases that is on par with world leading experts. We’re now working to develop something that can be deployed at scale around the world. 

To be successful, AI systems have to be trained on very diverse data. Due to the huge diversity and number of people served by the NHS, our data sets are really perfect for the development of AI systems. I think Moorfields, the NHS and the UK are the place to be for the most cutting-edge research in the field of AI and healthcare.