DeepMind is at it again. This time the company is using its powerful artificial intelligence software to look for various diseases in the eyes, according to a new paper seen by the Financial Times.
A key part of paper suggests that the AI may be better at spotting symptoms of disease than humans -- both in terms of accuracy, as well as speed.
DeepMind has been working with Moorfields Eye Hospital, Islington, to examine retinal scans of glaucoma, diabetic retinopathy and age-related macular degeneration. The paper, which has been submitted to a peer-reviewed medical journal, has showed "promising signs," and DeepMind hopes that its technology could enter clinical trials in a few years' time.
The clinical lead at DeepMind, Dominic King, told the Financial Times: "In specific areas like medical imaging, you can see we're going to make really tremendous progress in the next couple of years with artificial intelligence. Machine learning could have a very important role picking up things more sensitively and specifically than currently happens."
DeepMind trained a machine learning algorithm to look at thousands of retinal scans and recognise symptoms of each of the three diseases, then diagnosing it quicker and more effectively than humans could. This means that, in time, the AI could take some of the repetitive tasks away from the NHS and doctors and do it itself, giving doctors more time to care for patients.
DeepMind has been working with Moorfields Eye Hospital to analyse thousands of retinal scans, meaning it can identify various eye diseases accurately. (Image: osde8info)
According to the Financial Times, there are plans for DeepMind's technology to be used at University College London Hospitals to look at radiotherapy scans and at Imperial College London to analyse mammograms. It seems as though, having beaten two incredibly complex board games with its AI, DeepMind is now turning its attentions to the real-world, with health at the top of the agenda.
Company CEO Nik Storonsky is convinced that his company is an acquisition target because of its ability to greatly speed up the time it takes to detect credit card theft and other types of financial fraud.