fashion -WSS

fashion

author：leisure time 2025-10-02 21:07:22 - browse：5

Blue numbers and letter falling down a tunnel. -- health tech coverage from STAT — Adobe

A growing number of AI tools are being used to predict everything from sepsis to strokes, with the hope of accelerating the delivery of life-saving care. But over time, new research suggests, these predictive models can become a victim of their own success — sending their performance into a nosedive and generating inaccurate, potentially harmful results.

“There is no accounting for this when your models are being tested,” said Akhil Vaid, an instructor of data-driven and digital medicine at the Icahn School of Medicine at Mount Sinai and author of the new research, published Monday in the Annals of Internal Medicine. “You can’t run validation studies, do external validation, run clinical trials — because all they’ll tell you is that the model works. And when it starts to work, that is when the problems will arise.”

Vaid and his Mount Sinai colleagues simulated the deployment of two models that predicted a patient’s risk of dying and acute kidney injury within five days of entering the ICU. Their simulations assumed that the models did what they were supposed to — lower deaths and kidney injury by identifying patients for earlier intervention.

Get unlimited access to award-winning journalism and exclusive events.

Subscribe Log In

Previous article： After affirmative action ruling, medical educators look to 'holistic review'
Next article： What is STAT's opinion section for?

focus

hotspot

WSS

STAT's 2023 biotech VC rankings: ARCH Venture Partners tops list

FDA cancer chief on how to get things done: ‘Don’t tell anybody'

Stanford updates investigation of president’s past research

Trump got a 'target letter.' What is one and what does it mean?

Medical experts are missing from AI regulatory conversations

How dance can help health care workers heal from trauma

hotspot

fashion

entertainment

Dobbs anniversary: the lost opportunity of abortion as health care

Seattle police shoot person suspected in downtown stabbing

New air purifier tech could be key to improving indoor air quality

Why some people catch Covid but never get sick

Medicare proposal to cover medical devices draws a mixed reaction

What is STAT's opinion section for?

FDA approval of Sarepta Duchenne genetic therapy gives me hope

Experimental asthma app from Apple and Anthem shows promise

23andMe had bad news about my health. I wish a person had delivered it

DOJ, FTC to increase scrutiny of all types of health care mergers

hotspot

Plane’s emergency slide plummets from sky, hits Chicago home

DOJ, FTC to increase scrutiny of all types of health care mergers

Readout Newsletter: Ozempic panic wanes, Legend x Novartis, etc.

India sets sights on home

No Labels group has 'steep climb,' but third party 'worth trying': Larry Hogan

Readout Newsletter: Novo Nordisk trial, Verve base editing, etc

hotspot

focus

hotspot

explore