LibraryFamily 10 · Workflow & Operational Efficiency
Emergency department length-of-stay forecast
CuratedPilotEmerging evidence
Reviewed for completeness and published to the Library.
Predicts expected length of stay per patient at triage so the ED can adjust staffing and bed allocation in near-real time.
Plain-language summary
AIChoose an audience and generate a tailored summary on demand. The AI uses only what is on this page.
Clinical context
- Clinical problem
- Predict a future risk
- Point of care
- Triage
- Nature of AI output
- A risk score
- Clinical specialty
- Emergency medicine
- Care setting
- Emergency department
- Patient population
- Adults arriving at the emergency department.
- Intended use
- Predicts expected length of stay per patient at triage so the ED can adjust staffing and bed allocation in near-real time.
Technology
- AI technique
- Classical machine learning, Statistical model
- Input data
- Structured EHR data, Vital signs
- Output type
- Risk score
- Autonomy level
- Informs a human (advisory)
- Model provenance
- Built in-house
- Model version
- los-1.0
- Built on a general-purpose model
- No
Deployment
- Status
- Pilot
- Country
- Denmark
- Deployment date
- 15 January 2025
- Sites
- 1
Regulatory & governance
- EU AI Act risk tier
- Minimal risk
- High-risk basis
- Not applicable
- Medical device
- No
- EU MDR class
- Not a device
- CE marking
- Not required
- FDA status
- Not applicable
- ISO 14971 risk class
- Low
- GDPR processing basis
- Legitimate interest
- GDPR DPIA
- Not required
- Data identifiability
- Pseudonymised
- Explainability method
- Intrinsic
- Human oversight model
- Operations team treats the forecast as one input; admission decisions remain clinical.
NICE evidence standards
- ESF tier
- Tier A — system / service
- Evidence category
- Category 1
Performance summary
- Headline metric
- AUC / AUROC
- Value
- 0.79
- Subgroup performance assessed
- No
- Known bias signals
- —
Similar deployments
AILooking for close matches in the Library…
Evidence records
Studies and evaluations attached to this use case.
Retrospective validation
AUC / AUROC: 0.79
Population: Three-year ED-admission cohort, 60k visits
Contributors
- Deploying organisation
- [demo] Aarhus University Hospital · Hospital / health system · Denmark
- AI vendor
- —
- Product name
- —