Hybrid novice-AI system achieves expert-level performance in intraoperative ischemia detection

During carotid endarterectomy surgery, there is a risk of cerebral ischemia, which is typically monitored by specialists using continuous electroencephalography. Researchers developed a hybrid system combining novices with limited training and artificial intelligence to detect ischemia. The system dynamically combines inputs from novices and AI to produce a final result. A study with four novices showed that the hybrid system was equally effective as experts in sensitivity and false-positive rate. At 80% sensitivity, the hybrid system reduced false positives by half compared to the AI-only system. The results suggest that the hybrid system could be an effective tool for monitoring cerebral ischemia during surgery.