Spark Triage: Building an AI Assistant for Emergency Care

Description and use

Spark Triage is an AI‑powered triage assistant that processes patient complaints, applies deterministic rules, and presents urgency levels for clinician review, aiming to improve emergency department workflows while remaining adaptable to low‑resource settings.

Problem

Emergency departments need rapid triage; clinicians must assess multiple patients under pressure and decide urgency, but current processes lack AI support and deterministic rules integration.

Outcome

Spark Triage prototype demonstrates that an AI agent combined with deterministic triage rules can generate urgency proposals for emergency care, enabling clinician review and potential deployment in resource‑constrained settings.

Workflow

• Collect patient presenting complaint
• AI agent processes complaint and extracts supporting evidence
• Deterministic SATS rules calculate proposed urgency level
• System displays proposed urgency to clinician
• Clinician reviews, confirms or overrides the result
• Audit log records workflow actions

Evidence

Spark Triage is an AI-powered emergency triage desk assistant that helps organize patient information and generate proposed urgency levels for clinical review.

Source

The system explores how AI can help process presenting complaints, identify supporting evidence, and organize triage workflows.

Source

Rather than allowing a language model to make the final decision on its own, the system separates AI assistance from rule-based scoring.

Source