Forbes May 22, 2024
Victoria Forster

An FDA-approved AI algorithm was more likely to wrongly indicate the presence of cancer in Black women compared to White, Hispanic and Asian women, according to a new study.

“As more AI algorithms are becoming FDA approved for screening mammography, we were interested in knowing this performance was similar across all patient demographics and characteristics,” said Derek L. Nguyen, M.D., assistant professor at Duke University in Durham, North Carolina and lead author of the study. “The impact of patient characteristics on AI performance has not been well studied.”

The AI tool is FDA approved to help radiologists identify potentially concerning results on mammograms, which then leads to the patient being sent for further testing. There is some evidence that AI...

Today's Sponsors

Venturous
Got healthcare questions? Just ask Transcarent

Today's Sponsor

Venturous

 
Topics: AI (Artificial Intelligence), FDA, Govt Agencies, Patient / Consumer, Provider, Survey / Study, Technology, Trends
How To Get Comfortable With AI, Conversation With Andrew Ng
The great software rewiring: AI isn’t just eating everything; it is everything
The future of clinical trials: Are we ready?
AI Around The World – TSMC Pumps More Money Into U.S. Operations
DOJ: Google Can Keep AI Investments but Must Sell Chrome

Share This Article