South KoreaHealthcare and life sciences2023 – ongoingInha University Hospital
DeepCARS cardiac-arrest risk alerts on general wards
Inha's AI-alert study found cardiac-arrest rates of 1.06% with intervention versus 2.07% with usual care
Problem
General-ward staff needed to identify deteriorating patients early enough to intervene. Routine vital-sign measurements provided data for assessing that risk.
Approach
The hospital prospectively evaluated VUNO Med-DeepCARS during 2023, screening 36,797 general-ward admissions and including 35,627. The system used 4 vital signs to predict cardiac-arrest risk within 24 hours. Clinicians reviewed alerts and decided whether to reassess patients or provide additional treatment.
Results
Takeaway
The comparison concerned clinical responses after alerts, so the reported difference cannot be read as a hospital-wide before-and-after reduction caused by installing AI.
Categories
- Industry
- Healthcare and life sciences
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Vendor solution
- Goal
- Quality and risk
- Evidence
- Independent study
Sources
Source summaries
The peer-reviewed study compared clinical responses to AI alerts during 2023 and reported cardiac-arrest rates of 1.06% versus 2.07%. It was non-randomized and received funding from VUNO.
Full details are in the original sources.