Skip to Content
Artificial intelligence

DeepMind’s new project aims to prevent hospital deaths

February 26, 2018

Alphabet’s artificial-intelligence subsidiary DeepMind is partnering with the US Department of Veterans Affairs to try to predict when a patient is about to take a turn for the worse, even before any outward signs of getting sick.

Life or death: Roughly 11 percent of all in-hospital deaths occur because someone misses or fails to act properly on warnings signs that a patient’s condition is declining.

How it will work: The project will analyze patterns from 700,000 anonymized medical records to see if machine learning can identify which patients will get worse during a hospital stay and correctly predict the decline.

Practical applications: The VA and DeepMind say one possible use for the technology is spotting kidney failure or damage early on. Kidney injuries commonly occur after routine procedures and operations, like a hip replacement, and often come on suddenly with no obvious warning signs. These injuries can be life-threatening, so DeepMind wants to find ways to detect them earlier in hopes of saving patients.

Deep Dive

Artificial intelligence

What does GPT-3 “know” about me? 

Large language models are trained on troves of personal data hoovered from the internet. So I wanted to know: What does it have on me?

DeepMind has predicted the structure of almost every protein known to science

And it’s giving the data away for free, which could spur new scientific discoveries.

An AI that can design new proteins could help unlock new cures and materials 

The machine-learning tool could help researchers discover entirely new proteins not yet known to science.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.