In the US, sepsis affects at least 1.7 million adults annually, and about 350,000 of them pass away from the deadly blood infection that can set off a potentially fatal chain reaction that affects the entire body.
In the new study, researchers found that by utilizing a unique AI algorithm in emergency departments at UC San Diego Health, they were able to quickly predict sepsis infection in high-risk patients and reduce mortality by 17%. Image Credit: UC San Diego Health
To promptly identify patients at risk for sepsis infection, researchers at the University of California San Diego School of Medicine used an artificial intelligence (AI) model in the emergency departments at UC San Diego Health. Their findings were published in the online edition of npj Digital Medicine on January 23rd, 2024.
According to the research, the AI system called COMPOSER, which the researchers had previously created, reduced mortality by 17%.
Our COMPOSER model uses real-time data to predict sepsis before obvious clinical manifestations, it works silently and safely behind the scenes, continuously surveilling every patient for signs of possible sepsis.
Gabriel Wardi, Study Co-author and Chief, Division of Critical Care, Department of Emergency Medicine, UC San Diego School of Medicine
Upon arrival to the emergency room, the algorithm starts tracking over 150 patient characteristics, including test results, vital signs, current medications, demographics, and medical history, all of which may be associated with sepsis.
If a patient exhibits several factors that put them at high risk of developing sepsis, the AI system will alert nursing staff through the hospital's EHR. Following their review, the medical team and the nursing staff will decide on the best course of action.
These advanced AI algorithms can detect patterns that are not initially obvious to the human eye and the system can look at these risk factors and come up with a highly accurate prediction of sepsis. Conversely, if the risk patterns can be explained by other conditions with higher confidence, no alerts will be sent.
Shamim Nemati, Associate Professor, Department of Biomedical Informatics, UC San Diego School of Medicine
Shamim Nemati is also the Director of Predictive Analytics.
The study looked at over 6,000 patient admissions at the emergency rooms at Jacobs Medical Center in La Jolla and UC San Diego Medical Center in Hillcrest, both before and after COMPOSER was implemented.
It is the first study to show that using an AI deep-learning model that uses artificial neural networks as a check and balance to safely and accurately identify patient health issues improves patient outcomes. The health care team reviews the identified complicated and multiple risk variables by the model for confirmation.
Gabriel Wardi said, “It is because of this AI model that our teams can provide life-saving therapy for patients quicker.”
Wardi is also an emergency medicine and critical care physician at UC San Diego Health.
After being turned on in December 2022, COMPOSER is currently used in numerous hospital in-patient units run by UC San Diego Health. The newest facility of the health system, UC San Diego Health East Campus, will shortly activate it.
The first academic medical system in the area, UC San Diego Health, is leading the way in AI health care. It recently announced the appointment of its first chief health AI officer. It opened the Joan and Irwin Jacobs Center for Health Innovation, which aims to create cutting-edge medical solutions.
Furthermore, the health system started a pilot program in which ChatGPT automatically generates more sympathetic message responses based on Microsoft's generative AI integration with Epic, a cloud-based electronic health record system. This eliminates the need for doctors and caregivers to perform this extra step and frees them up to concentrate on patient care.
Integration of AI technology in the electronic health record is helping to deliver on the promise of digital health, and UC San Diego Health has been a leader in this space to ensure AI-powered solutions support high reliability in patient safety and quality health care.
Christopher Longhurst, MD, Executive Director, Jacobs Center for Health Innovation, UC San Diego Health
Christopher Longhurst is also a Chief Medical Officer and Chief Digital Officer at UC San Diego Health.
The study was co-authored by Aaron Boussina, Theodore Chan, Allison Donahue, Robert El-Kareh, Atul Malhotra, Robert Owens, Kimberly Quintero, and Supreeth Shashikumar, all at UC San Diego.
The research was sponsored, in part, by the National Institutes of Health, the National Library of Medicine, and the Joan and Irwin Jacobs Center for Health Innovation at UC San Diego Health.
Journal Reference:
Boussina, A., et.al., (2024). Impact of a deep learning sepsis prediction model on quality of care and survival. npj Digital Medicine. doi.org/10.1038/s41746-023-00986-6.
Source: https://health.ucsd.edu/