New TriageSim framework generates synthetic emergency triage conversations from EHR data

Researchers have developed TriageSim, a framework designed to generate realistic conversational simulations for emergency triage scenarios. This system leverages structured electronic health records to create persona-conditioned dialogues, complete with synthetic transcripts and audio. The generated corpus, comprising approximately 800 synthetic conversations, has been evaluated for linguistic, behavioral, and acoustic fidelity, with a subset also assessed for medical accuracy. The framework's utility is demonstrated through its application in conversational triage classification, showing modest agreement across different modalities. AI

IMPACT Enables more robust training and evaluation of AI systems for medical dialogue and triage.

RANK_REASON The cluster describes a research paper published on arXiv detailing a new simulation framework. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Dipankar Srirag
• electronic health records
• Gotit.pub
• Hugging Face
• ScienceCast
• TriageSim

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →