sciencewr.com

Hospital Computer Predicts Death (VIDEO)

First Posted: Sep 15, 2015 11:20 AM EDT
Close

The Beth Israel Deaconess Medical Center in Boston has unveiled a computer that can do the unthinkable: predict the death of human patients with almost 100 percent accuracy.

The supercomputer is linked to every patient in the center, and contains information from more than 250,000 patients over the past 30 years. The longer it's in use, the more information it gathers, and the more accurate it becomes at recognizing illnesses.

"Our goal is not to replace the clinician. This artificial intelligence is really about the augmenting of doctors' ability to take care of patients," Dr. Steve Horng told BBC.

The computer can sometimes accurately suggest what is wrong with a patient before a doctor has made their diagnosis, an extremely powerful aid in the diagnostic process, according to Mirror.

It takes measurements of blood pressure and oxygen levels every three minutes, staying very much on top of the patients it is linked to. This allows for the computer to predict the likelihood of specific diseases.

"So, for example this computer thinks that this patient has a 99% chance of having some sort of chest pain and they only have 26% of having heart failure," Horng said to the Huffington Post.

The computer has proved to be highly accurate thus far, and has greatly helped the doctors in the medical center.

"We can predict with 96% confidence when patients [are facing a high] probability of dying," Horng said. "If the computer says you're going to die, you probably will die in the next 30 days."

Related Articles

New Computer Program Recognizes Sketches Better Than Humans

For more great science stories and general news, please visit our sister site, Headlines and Global News (HNGN).

See Now: NASA's Juno Spacecraft's Rendezvous With Jupiter's Mammoth Cyclone

©2017 ScienceWorldReport.com All rights reserved. Do not reproduce without permission. The window to the world of science news.

Join the Conversation

Real Time Analytics