United StatesHealthcare and life sciences2019 – ongoingJohns Hopkins Health System
TREWS early warning for sepsis
Johns Hopkins reported that TREWS identified more than 80% of patients with sepsis before onset
Problem
Early sepsis was difficult to diagnose because its symptoms overlapped with other illnesses. Existing alert criteria produced frequent false warnings.
Approach
Johns Hopkins computer scientist Suchi Saria and colleagues developed TREWS to assess patient-record data for sepsis risk. The system alerted clinicians through electronic health records and prompted clinical responses. It was operating at Howard County General Hospital and Suburban Hospital by June 2019.
Results
Takeaway
TREWS delivered warnings within clinicians' existing records. Its reported detection rate did not by itself measure an improvement in patient survival.
Categories
- Size
- Large enterprise
- Industry
- Healthcare and life sciences
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Built in-house
- Goal
- Quality and risk
- Evidence
- Company's own claim
Sources
Source summaries
Johns Hopkins described TREWS operating at 2 hospitals and planned expansion to other facilities. A researcher reported that it detected more than 80% of patients with sepsis before onset.
Full details are in the original sources.