AI Spots Infections in Post-Surgery Wound Photos

Mayo Clinic’s system offers remote monitoring potential

An artificial intelligence (AI) system developed by **Mayo Clinic** researchers can now analyze patient-submitted photos of post-operative wounds, accurately detecting surgical site infections (SSIs), potentially speeding up both reassurance and treatment.

Key Development

The AI system, described in Annals of Surgery, was trained using over 20,000 images gathered from more than 6,000 patients across nine **Mayo Clinic** hospitals.

The system first checks if an image contains a surgical incision, then assesses the image quality, and finally, looks for signs of infection.

Cornelius Thiels, D.O., a surgical oncologist at **Mayo Clinic**, explained the motivation behind the technology: “We were motivated by the increasing need for outpatient monitoring of surgical incisions in a timely manner.”

The AI model achieved 94% accuracy in identifying incision presence and an area under the curve (AUC) of 0.81 in detecting infections, while maintaining consistent performance across diverse patient demographics, mitigating bias concerns.

The CDC estimates that SSIs occur in 2-4% of patients undergoing inpatient surgeries in the U.S. (CDC).

Future Applications

The research team envisions integrating the tool into real-world surgical care workflows.

“For patients, this could mean faster reassurance or earlier identification of a problem,” notes **Hala Muaddi**, M.D., Ph.D., a fellow at **Mayo Clinic**.

The team hopes the technology will support patients recovering at home and act as a screening tool, alerting physicians to potentially concerning incisions.

Hojjat Salehinejad, Ph.D., a consultant at the **Kern Center** for the Science of Health Care Delivery, added, “Our hope is that the AI models we developed — and the large dataset they were trained on — have the potential to fundamentally reshape how surgical follow-up is delivered.”

Reference:Hala Muaddi, Choudhary A, Lee F, et al. Imaging Based Surgical Site Infection Detection Using Artificial Intelligence. Ann Surg. 2025. doi: 10.1097/sla.0000000000006826