Google’s AI Can Predict When a Hospital Patient Will Pass Away

By: | July 16th, 2018

Tech giant Google has already made inroads into nearly every aspect of people’s lives. And now, Google might even know when you’re going to die. But, ominous as it sounds, this information could end up saving lives.

Google’s Medical Brain team has created a new type of artificial intelligence algorithm to predict death among hospital patients.

A paper published in the journal Nature detailed how Google’s predictive algorithm could determine whether inpatients will survive, using a vast amount of medical data from two U.S. hospitals. It was 95 percent accurate in the first hospital and 93 percent accurate in the second hospital while predicting inpatient mortality.

A woman with metastatic breast cancer was given a 19.9 percent chance of dying in the hospital by the Medical Brain algorithm. (Google’s AI analyzed 175,639 data points from the woman’s medical records in the process.) The hospital’s Early Warning Score gave her a 9.3 percent chance of dying. Unfortunately, the Google algorithm was right: The patient died within two weeks.

Google research scientist Alvin Rajkomar said, “When patients get admitted to a hospital, they have many questions about what will happen next.

“When will I be able to go home? Will I get better? Will I have to come back to the hospital?

“Predicting what will happen next is a natural application of machine learning.”

AI can also predict unplanned re-admissions within 30 days and probable length of stay at a hospital. Google believes that the system could be used to predict death far longer in advance in the future.

Nidhi Goyal

Nidhi is a gold medalist Post Graduate in Atmospheric and Oceanic Sciences.

More articles from Industry Tap...