Skip to main content

Google’s DeepMind will soon apply artificial intelligence to the detection of eye diseases

Oops! We couldn't load this video player
DeepMind, the London-based artificial intelligence lab acquired by Google in 2014, has accomplished more than a few spectacular stunts of machine learning. Its neural networks bested a human champion at the notoriously tough game of Go, inculcated the basic rules of soccer on a digital ant-like creature, and teased out winning strategies for more than 49 Atari 2600 games. But now, the outfit’s robots are being tasked with a more humanistic pursuit: eye disease research.

On Tuesday, DeepMind announced a long-term project that will see the company’s machine-learning algorithms parse “millions” of eye scans to tease out early warning signs that human doctors might otherwise miss.

Recommended Videos

The new project, which is based out of the U.K.’s Moorfields Eye Hospital in east London, is the fruit of DeepMind’s ongoing partnership — dubbed DeepMind Health — with the country’s National Health Service. In February, the firm launched Streams, an iPhone app for the Royal Free Hospital in London which helps to inform doctors and nurses of kidney patient complications that might interfere with particular sorts of treatment. And in March, it announced plans to launch Hark, a task management start-up that it acquired earlier this year, in London hospitals.

Please enable Javascript to view this content

But the Moorsfields program is the first to leverage DeepMind’s machine-learning smarts, and also the first to focus on “purely medical research.”

The firm’s computers will analyze millions of retinal scans collected from the institute’s patients. DeepMind’s researchers will then use those scans to train algorithms to identify early warning signs for chronic diseases like macular degeneration and diabetic retinopathy, according to DeepMind co-founder Mustafa Suleyman.

The impetus was a request by Pears Keane, a consultant opthalmologist at Moorsfields who became intrigued by the health implications of artificial intelligence after reading about DeepMind’s early successes with Atari video games. “I’d been reading about deep learning and the success that technology had in image recognition,” Keane told The Guardian. “Within a couple of days I got in touch with [Suleyman], and he replied.”

The “incredibly detailed” records provide invaluable source material for the outfit’s neural networks, Suleyman told The Guardian. “There’s so much at stake, particularly with diabetic retinopathy,” he said. “If you have diabetes you’re 25 times more likely to go blind. If we can detect this, and get in there as early as possible, then 98 percent of the most severe visual loss might be prevented.”

Professor Peng Tee Khaw said the collaboration, which has the support of the royal National Institute of Blind people (RNIB) and charities such as the Macular Society, is not necessarily intended to remove doctors from the prognostic process. Rather it is to make their day-to-day jobs easier by producing yet another resource upon which they can draw — a body of medical insights that might otherwise take years to compile.

“It takes me my whole life experience to follow one patient’s history. And yet patients rely on my experience to predict their future,” he told The Guardian. “If we could use machine-assisted deep learning, we could be so much better at doing this, because then I could have the experience of 10,000.”

Artificial neural networks like the products of DeepMind’s research have already shown great promise in the field of healthcare. NVIDIA recently announced a partnership with Massachusetts General Hospital to apply artificial intelligence techniques to the detection and treatment of diseases, and one neural network model, the subject of a 2012 study, correctly identified coronary artery disease with 91.2-percent accuracy (others have isolated symptomatic patterns in diseases from cancer to diabetes).

“[Machine learning can] improve patient safety, quality of care, reduce medical costs and save lives,” wrote former UW Medicine Health System researcher Peter Ghavami.

But their application has raised privacy concerns. After it was revealed that DeepMind Health would grant the Google-owned firm uninhibited access to 1.6 million NHS patient records from London’s Royal Free Hospital, Chase Farm, and Barnet hospitals over the past five years and until 2017, some privacy advocates protested. The Information Commissioner’s Office, the U.K’s data protection watchdog, began investigating the arrangement in May.

DeepMind, for its part, said that any sensitive data in the course of research is being transmitted securely. “As Googlers, we have the very best privacy and secure infrastructure for managing the most sensitive data in the world,” Suleyman told The Guardian. But it also emphasized that participation in the program is voluntary.

“Patients can opt out of any data-sharing system by emailing the Trust’s data protection officer,” according to a Q&A on Moorfield’s website.

Kyle Wiggers
Former Digital Trends Contributor
Kyle Wiggers is a writer, Web designer, and podcaster with an acute interest in all things tech. When not reviewing gadgets…
This iBuyPower gaming PC with RTX 4060 is under $1,000 — for now
The iBUYPOWER Trace 7 Mesh gaming desktop on a white background.

Gaming PC deals worth buying still usually cost more than $1,000 after the discounts, but here's an offer from Best Buy that's available for a more affordable price. The iBuyPower Trace 7 Mesh, which is originally sold for $1,300, is down to just $900 following a $400 discount. We're not sure how much time is remaining before this bargain ends, so if you're interested in this gaming desktop, you need to push forward with your purchase immediately if you want to secure the savings.

Why you should buy the iBuyPower Trace 7 Mesh gaming PC
The iBuyPower Trace 7 Mesh is a relatively affordable gaming PC, but it doesn't sacrifice much in terms of performance. It runs on the AMD Ryzen 7 5700 processor and the Nvidia GeForce RTX 4060, which is in our list of the best graphics cards as our recommendation for 1080p gaming. It has 16GB of RAM, which is the best place to start for a gaming PC, according to our guide on how much RAM do you need. With these components, you won't have trouble playing the best PC games, though you'll have to go with medium settings for the more demanding titles.

Read more
Google gives memory superpowers to Gemini for more natural chats
Google Gemini running on an Android phone.

Google is finally bringing a crucial new feature to Gemini that will solve a key pain point of interacting with its AI chatbot. The company is enabling a memory feature which allows Gemini to pull up details from a past conversation.

“Whether you’re asking a question about something you’ve already discussed, or asking Gemini to summarize a previous conversation, Gemini now uses information from relevant chats to craft a response,” says a Google update.

Read more
Apple is working on a mini-LED Studio Display, but the iMac Pro is nowhere to be found
A person using a Mac Studio with an Apple Studio Display.

If you’ve been craving a souped-up Studio Display that improves on the original model released in 2022, there’s been some good news: it looks like a new model with a mini-LED panel and a ProMotion 120Hz refresh rate is on the way. Yet it’s left me with more questions than answers about Apple’s long-lost iMac Pro.

In the world of Apple leaks, you quickly learn that there are two sources whose reliability stands above the rest: Bloomberg reporter Mark Gurman and display industry expert Ross Young. When these two agree on something, it’s pretty much nailed on, and that’s exactly what just happened regarding Apple’s Studio Display.

Read more