When It Comes to Health Care, AI Has a Long Way to Go

That’s because health data such as medical imaging, vital signs, and data from wearable devices can vary for reasons unrelated to a particular health condition, such as lifestyle or background noise. The machine learning algorithms popularized by the tech industry are so good at finding patterns that they can discover shortcuts to “correct” answers that won’t work out in the real world. Smaller data sets make it easier for algorithms to cheat that way and create blind spots that cause poor results in the clinic. “The community fools [itself] into thinking we’re developing models that work much better than they actually do,” Berisha says. “It furthers the AI hype.”


→ Continue reading at WIRED

Similar Articles


Most Popular

'Fall Guys' lands on Switch, Xbox and Epic Games Store on June 21st

It's been over a year since Mediatonic confirmed Fall Guys: Ultimate Knockout was coming to Xbox and Nintendo Switch. After some delays, the wait is...

The Morning After: Apple may be testing USB-C iPhones

Oh my, it might actually be happening. I’ve wanted this for a while, and Apple’s excuses for sticking with its Lightning connector get weaker...

Suspects in dozens of Seattle break-ins, burglaries charged

Branden Cerna and Stanley Lee face a number of charges, including first-degree theft and theft of a motor vehicle. SEATTLE — A pair...