A research team led by Dr. Nor-Eddine Regnard, a musculoskeletal radiologist and chief medical officer at AI software developer Gleamer.ai, retrospectively applied the company's AI BoneView software to all x-rays performed for a suspected limb or pelvic bone fracture over a three-month period at a group of 17 private imaging centers in France. A senior skeletal radiologist then assessed all exams that had discrepant results from the original radiologist interpretation.
After analyzing the 4,774 exams, the AI software detected 235 additional fractures, corresponding to a 26% increase in fracture detection and a 5% error rate in the original radiologist interpretations. The researchers determined that knowledge of the fracture could have led to a change in therapeutic care for 144 (61.3%) of these fractures and a possible surgery for 51 (21.7%) fractures.
"AI-aid has the potential to prevent medical errors like missed fractures on initial x-ray exams, thus impacting patient care," the authors wrote.
Delve further into their results by viewing this presentation on Sunday.
