A Novel AI Algorithm With Enhanced Accuracy for Surveillance of Deep Surgical Site Infections After Elective Colorectal Surgery. A Diagnostic Accuracy Study.

What this Trial Is About

Healthcare-associated infections, including surgical site infections (SSI), are a major concern for patient safety and healthcare systems worldwide. Surveillance is a key component recommended by the World Health Organization to prevent these infections. This research aims to evaluate a new artificial intelligence algorithm designed to detect SSIs in patients who have undergone elective colorectal surgery, comparing it to traditional manual surveillance methods within a national infection surveillance program. The study compares two groups of patients undergoing colorectal surgery: one monitored using the standard manual method for identifying SSIs, and the other assessed with a novel AI algorithm. This algorithm analyzes clinical data such as clinical notes, microbiology reports, and coding for diagnoses and complications to identify suspected infections at three different anatomical levels. The surveillance covers up to 30 days after surgery to detect infections occurring after hospital discharge. Participants will be followed up with continuous monitoring to detect surgical site infections within the 30-day post-operative period. Researchers will assess the rate of SSIs detected by both the manual method and the AI algorithm. The study uses data routinely collected during postoperative care, focusing on how well the AI system identifies infections compared to current practices. The study is observational and does not involve treatment interventions, and participation duration depends on the postoperative surveillance period.

AI Algorithm for Surveillance of Deep Surgical Site Infections After Elective Colorectal Surgery.