Google predicts when people will die ‘with 95% accuracy’

Google has claimed it can predict with 95% accuracy when people will die using new artificial intelligence technology.

In a paper published in the journal Nature, the company’s Medical Brain team detailed how it is using a new type of artificial intelligence algorithm to make predictions about the likelihood of death among patients in two separate hospitals.

For predicting patient mortality, Google’s Medical Brain was 95% accurate in the first hospital and 93% accurate in the second.

Subscribe to The Week

Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.

SUBSCRIBE & SAVE

Sign up for The Week's Free Newsletters

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

Sign up

It works by analysing patient’s data, such as their age, ethnicity and gender. This information is then joined up with hospital information, like prior diagnoses, current vital signs, and any lab results, reports The Sun.

But according to Bloomberg, what impressed medical experts most “was Google’s ability to sift through data previously out of reach: notes buried in PDFs or scribbled on old charts. The neural net gobbled up all this unruly information then spat out predictions. And it did it far faster and more accurately than existing techniques.”

It is not the first time Google has made inroads into the medical industry. Its DeepMind subsidiary, considered by some experts to lead the way in AI research, “courted controversy” in 2013 after it was revealed it had access to 1.6 million medical records of NHS patients at three hospitals, reports The Independent.

Yet despite concerns the search giant could be given access to even more data, the latest findings prove Google could have a potentially life-saving impact on its 1.17 billion users worldwide.

The Medical Brain team said: “These models outperformed traditional, clinically used predictive models in all cases. We believe that this approach can be used to create accurate and scalable predictions for a variety of clinical scenarios.”

“For medical facilities bogged down in bureaucratic red tape, Google’s software is a godsend” says Vanity Fair. “Not only can it predict when a patient may die, but it can also estimate how long someone might stay in a hospital, or the chance they’ll be readmitted”.

But the magazine also offers a word of warning saying that “for patients, giving a tech giant like Google access to sensitive medical information may have unintended consequences”.