This neural network can diagnose pneumonia better than your doctor
This could save millions of lives

"Add diagnosing dangerous lung diseases to the growing list of things artificial intelligence can do better than humans," said Jackie Snow at MIT Technology Review. Researchers from Stanford University have developed a neural network that can "beat radiologists at spotting pneumonia" on chest X-rays. When the network's algorithm was expanded, it also proved better than doctors at identifying 13 other diseases. More than 1 million adults are hospitalized with pneumonia each year in the U.S.
The illness is also "the single largest infectious cause of death" for children worldwide, so early detection could save millions. Radiologists and other physicians are "on notice," say the researchers. Other AI experiments have produced algorithms that can "identify heart arrhythmias better than a human expert" and spot cancerous skin lesions "just as well as a board-certified dermatologist."
A free daily email with the biggest news stories of the day – and the best features from TheWeek.com
The Week
Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.

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.
-
Is this the end of ultraprocessed foods?
Today's Big Question California law, MAHA movement on the same track
-
Are inflatable costumes and naked bike rides helping or hurting ICE protests?
Talking Points Trump administration efforts to portray Portland and Chicago as dystopian war zones have been met with dancing frogs, bare butts and a growing movement to mock MAGA doomsaying
-
Gaza’s reconstruction: the steps to rebuilding
Even the initial rubble clearing in Gaza is likely to be fraught with difficulty and very slow