A team of researchers from the Mayo Clinic has created an artificial intelligence (AI) system capable of detecting surgical site infections (SSIs) using wound images submitted by patients.

Currently, clinicians manually check patient-submitted wound images using online portals for SSI and other issues, with early identification of SSI specifically critical for reducing morbidity. However, this approach can be time consuming.

As such, the AI system has been designed to automatically identify surgical incisions, assess image quality and flag signs of infection in those photos submitted by patients via the online portals.

To create the system, the team trained it on roughly 20,000 images from more than 6,000 patients from nine Mayo Clinic hospitals. When trialed, the model detected surgical incisions with a 94% rate of accuracy as well as an 81% area under the curve in identifying infections.

“The AI model is based on deep learning, which uses layers of artificial neurons for abstract representation of input images. These layers extract features from patient-submitted images — such as edges and patterns — which may or may not be visible to the human eye but have information that can distinguish images based on different targeted outcomes,” the researchers noted.

Also capable of categorizing images and identifying low-quality images, the AI system promises to help reduce delays in diagnostics, thereby leading to earlier treatment of infections, and to offer improved care for patients recovering from surgeries at home.

“This work lays the foundation for AI-assisted postoperative wound care, which can transform how postoperative patients are monitored,” the researchers added. “For patients, this could mean faster reassurance or earlier identification of a problem. For clinicians, it offers a way to prioritize attention to cases that need it most, especially in rural or resource-limited settings.”

The developers suggest that the model could also be used on patient-submitted images of drains and ostomies, for instance.

An article detailing the team’s work, “Imaging Based Surgical Site Infection Detection Using Artificial Intelligence,” appears in the journal the Annals of Surgery.