AI radiology algorithm 19% more accurate than physicians

When a new machine learning-based artificial intelligence tool was used to classify hip fractures, it was right 92 percent of the time, The Washington Post reported Feb. 22.

Advertisement

A national shortage of radiologists paired with a rising demand for their services from an aging population means that radiologists are under pressure. Delays to surgery or treatment in hip fracture cases specifically can spell an increase in pressure sores or even death, so access to reliable and accurate care is vital.

A team of researchers in the U.K. trained an AI tool to identify hip joints and fractures from radiographs. Physicians were asked to classify more than 3,600 images of hip fractures, but the machine learning algorithm showed better accuracy in results. Overall, it made the correct diagnosis 92 percent of the time, as opposed to the physicians’ 77.5 percent of the time, making the machines 19 percent more accurate than the humans.

At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.

Register to Attend Webinar

How Allina Health is building the operating system for AI transformation — and driving $7.3M in added annual surgical contribution margin

Wednesday, July 29
1:00 PM - 2:00 PM CDT

Presenters: William Evans, Allina HealthDonna Sajady, Allina HealthDavid Atashroo, MD, Qventus

Advertisement

Next Up in Digital Health

Advertisement

Comments are closed.