用医疗算法构建问答图，让大模型更高效地完成问诊。
Using Medical Algorithms for Task-Oriented Dialogue in LLM-Based Medical Interviews
- 将医疗指南转为问答图谱，实现结构化问诊
- 冷启动机制降低初始提问负荷，响应后自适应调整路径
- 医生端生成结构化报告，减轻认知负担，适合临床使用
我们构建了一个基于有向无环图（DAG）的面向任务对话框架，用于大模型驱动的医疗问诊。系统包含：(1) 将医疗算法与指南转化为临床问题语料库的系统性流程；(2) 基于分层聚类的冷启动机制，无需患者信息即可高效启动提问；(3) 可扩展与修剪的动态分支与回溯机制，根据患者回答自适应调整路径；(4) 终止逻辑确保在收集足够信息后结束访谈；(5) 自动合成符合临床工作流的医生友好型结构化报告。人机交互原则指导了患者与医生端应用的设计。初步评估由五名医生使用标准化量表进行：NASA-TLX（认知负荷）、系统可用性量表（SUS）、用户界面满意度问卷（QUIS）。患者端表现优异：认知负荷低（NASA-TLX = 15.6），可用性高（SUS = 86），满意度强（QUIS = 8.1/9），尤其在易学性和界面设计上得分突出。医生端认知负荷中等（NASA-TLX = 26），可用性极佳（SUS = 88.5），满意度为8.3/9。两项应用均有效融入临床流程，降低认知负担并支持高效报告生成。局限包括偶发延迟及评估样本小且多样性不足。
原文摘要 · Abstract (English)
We developed a task-oriented dialogue framework structured as a Directed Acyclic Graph (DAG) of medical questions. The system integrates: (1) a systematic pipeline for transforming medical algorithms and guidelines into a clinical question corpus; (2) a cold-start mechanism based on hierarchical clustering to generate efficient initial questioning without prior patient information; (3) an expand-and-prune mechanism enabling adaptive branching and backtracking based on patient responses; (4) a termination logic to ensure interviews end once sufficient information is gathered; and (5) automated synthesis of doctor-friendly structured reports aligned with clinical workflows. Human-computer interaction principles guided the design of both the patient and physician applications. Preliminary evaluation involved five physicians using standardized instruments: NASA-TLX (cognitive workload), the System Usability Scale (SUS), and the Questionnaire for User Interface Satisfaction (QUIS). The patient application achieved low workload scores (NASA-TLX = 15.6), high usability (SUS = 86), and strong satisfaction (QUIS = 8.1/9), with particularly high ratings for ease of learning and interface design. The physician application yielded moderate workload (NASA-TLX = 26) and excellent usability (SUS = 88.5), with satisfaction scores of 8.3/9. Both applications demonstrated effective integration into clinical workflows, reducing cognitive demand and supporting efficient report generation. Limitations included occasional system latency and a small, non-diverse evaluation sample.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。