Digital Health
医疗降级沟通技能的自动化评估
Automated assessment of healthcare de-escalation communication skillsA method for evaluating verbal de-escalation competencies in nursing students using conversational AI and virtual reality simulations, comparing deep learning models (BERT, RoBERTa, XLNet) with a rubric-driven framework (RATAS). / 一种利用对话式人工智能和虚拟现实模拟评估护理学生口头降级能力的评估方法，比较了深度学习模型（BERT、RoBERTa、XLNet）与基于评分细则的框架（RATAS）。
- Knowledge type
- 实际应用
- First proposed or discovered
- Proposer
- Turkey Farm member
- Creator
- 农场主
- Last updated
- 2026-11-02 17:06:36
Sub-knowledge points
Speech-to-text conversion of spoken responsesComparison of fine-tuned deep learning modelsEvaluation of rubric-driven framework (RATAS)
Higher-order knowledge points
Digital health training environmentsConversational AI virtual reality simulationNursing education
Main parameters and performance indicators
De-escalation competencies: 11 specific skillsModels: BERT, RoBERTa, XLNet, RATASAccuracy metric: 0.9927 (RoBERTa validation)Dataset types: Controlled validation vs. Real-world student responses
Open problems
Performance degradation on real-world responses compared to validation datasetsStability of rubric-driven LLM-based assessment (RATAS) across different conditions
Related literature
ranked by community starsAutomated assessment of healthcare de-escalation communication skills in a conversational AI virtual reality simulationOpen source
Related products and links
No item recorded yet
Creation and modification trail
1 recordsRecord title: Automated assessment of healthcare de-escalation communication skills.
Node update feedback
review required