February 12, 2020
Five years after Google acquired DeepMind, the health and artificial intelligence group is unveiling its biggest breakthrough yet in health care. Its technology is able to predict if a patient has potentially fatal kidney injuries 48 hours before many symptoms can be recognized by doctors.
In a paper published on Wednesday in the journal Nature, DeepMind researchers said their algorithms correctly predicted 90 percent of acute kidney injuries that would end up requiring dialysis. The work was the result of a project with the U.S. Department of Veteran Affairs to help doctors get a head start on treatment.
“We’ve been really excited for the potential of using AI to support clinicians moving care from reactive to proactive and preventative,” said Dominic King, DeepMind’s co-founder and clinical lead, in an interview.
About 2 million people die every year across the globe from acute kidney injury, according to researchers from the University of Pittsburgh School of Medicine. The condition, which involves a sudden episode of kidney failure or damage, can be tricky for doctors to diagnose because there aren’t always immediate and clear symptoms. Studies have shown that catching it early can decrease the likelihood of serious injury or death.
In 2014, Google acquired DeepMind for a reported $500 million as it looked to expand in AI and bring in top industry experts to work on hard problems involving machine learning. As Alphabet and its various units have stepped into the health-care space in the past few years, much of the focus has been on using its technology to predict serious health outcomes before they happen.
DeepMind’s health projects will soon be folded into Google Health, led by David Feinberg. The group hasn’t said much publicly beyond its website, which says it’s studying how AI can be used to assist in “diagnosing cancer, predicting patient outcomes, preventing blindness, and much more.” Much of its team remains based in the U.K., although its health unit is expected to relocate to Google’s Silicon Valley headquarters.
Even in its early days, the company’s work in health care has been criticized for not adequately protecting user privacy. In a recent case, a patient sued Google and the University of Chicago Medical Center for not removing doctors’ notes and date stamps from personal medical records. And a U.K. government privacy watchdog said a hospital had illegally sent 1.6 million records to Google DeepMind for a new health-care app.
The research on kidney injuries came from two separate joint studies with the VA and the Royal Free Hospital in London. DeepMind said it analyzed data stored electronically from more than 100 VA hospitals, reviewing information on hundreds of thousands of patients. Personal details like names and social security numbers were stripped from the data.
In addition to predicting acute kidney disease two days early, the company is also researching how to deliver these alerts in emergency situations so doctors properly recognize and act on them.
DeepMind’s King said there’s still work to be done to create a regulatory framework for bringing predictive tools to medicine and to better understand how they can be delivered in real time.
DeepMind’s breakthroughs might eventually augment the mobile app Streams, which is mostly used in the U.K. as a communications tool by doctors and nurses. It doesn’t currently use AI, but DeepMind has long stressed its vision of someday building an “an AI-powered assistant for nurses and doctors everywhere.”