When a patient develops sepsis, every minute can matter. Recognizing the condition early and beginning appropriate treatment can be challenging, particularly because the warning signs may initially resemble other medical problems. At UC San Diego Health, Dr. Gabriel Wardi is helping explore how artificial intelligence can give physicians an additional tool for identifying and treating seriously ill patients.
Dr. Wardi is an emergency medicine and critical care physician and an associate professor at UC San Diego School of Medicine. He serves as Chief of the Division of Emergency Critical Care and works directly with patients in both emergency departments and intensive care units. His training combines biomedical engineering, public health, emergency medicine and critical care, giving him a multidisciplinary foundation for his work in medical technology and patient care.
One of the most notable projects associated with Dr. Wardi is COMPOSER, an artificial-intelligence system developed to identify patients at risk of sepsis. The system continuously evaluates more than 150 variables, including vital signs, laboratory results, medications, demographics and medical history. When the algorithm identifies a concerning pattern, it can alert the clinical team so physicians and nurses can evaluate the patient and determine appropriate treatment.
A 2024 UC San Diego study examining the technology found that implementation of COMPOSER was associated with a 17% reduction in sepsis-related mortality. More than 6,000 patient admissions were examined before and after the system was deployed in two UC San Diego emergency departments. The technology has since expanded to additional inpatient units.
Dr. Wardi’s recent work is taking the use of AI even further. In a study reported by UC San Diego in June 2026, researchers explored how large language models could assess the quality of sepsis care much faster than traditional manual chart reviews. The system can examine large numbers of medical charts and identify information relevant to complex sepsis quality measures, potentially providing physicians with useful feedback much closer to the time of treatment.
For Dr. Wardi, the objective isn’t to replace physicians with technology. Instead, his work focuses on putting useful information into clinicians’ hands at the right moment. His approach compares rapid clinical feedback to the way athletes receive performance statistics shortly after competing: timely information can help people understand what worked and where improvements can be made.
His research interests extend beyond sepsis to predictive monitoring, cardiac arrest, resuscitation and other challenges involving critically ill patients. His laboratory also investigates emerging technologies and predictive models that could improve care for patients before their conditions become more difficult to manage.
Dr. Wardi’s career demonstrates the increasingly important role of physician-scientists who understand both medicine and technology. By combining hands-on emergency and critical care with research into artificial intelligence, he is helping explore new ways of making hospitals safer and more responsive.
As healthcare continues to embrace AI, physicians like Dr. Wardi are positioned at an important intersection: ensuring that innovative technology remains connected to the realities of patient care. His work at UC San Diego represents a promising example of how physician-led innovation can turn advanced computing into practical tools designed to support better outcomes.
