Rochester, Minn.-based Mayo Clinic researchers have developed an AI tool to accurately identify signs of surgical site infection on postoperative online portal images, according to a study published July 3 in the Annals of Surgery.
“This work lays the foundation for AI-assisted postoperative wound care,” Hala Muaddi, MD, PhD, first author of the study and a hepatopancreatobiliary fellow at Mayo Clinic, said in a July 7 news release from the health system. “It’s especially relevant as outpatient operations and virtual follow-ups become more common.”
Here are three things to know about the tool:
- The tool was trained on more than 20,000 images from more than 6,000 patients across nine Mayo Clinic hospitals.
- After a patient uploads a postoperative wound photo to an online portal, the tool — called Vision Transformer — identifies whether the image contains a surgical incision and if the incision shows signs of infection.
The model was 94% accurate in detecting incisions and achieved an 81% area under the curve in identifying infections, and it performed consistently across diverse patient populations. - “The researchers are hopeful that this technology could help patients receive faster responses, reduce delays in diagnosing infections and support better care for those recovering from surgery at home,” the release said. “With further validation, it could function as a frontline screening tool that alerts clinicians to concerning incisions.”
Read the full study here.
