Jev 在实时临床问诊中辅助鉴别诊断
在实时临床问诊中,环境记录工具将转录文本周期性提供给 Jev,Jev 遍历医学本体,实时分类症状、更新鉴别诊断、建议治疗方案并识别危险信号。
作者分享将医疗接诊分诊的决策步骤从LLM迁移到Jev(@typesafeai)的实践:25个类型化问题、0.29秒、$0.00015,由代码驱动工作流,并附方法、数据与局限。
Healthcare intake is the hard case for AI triage: complex workflow, safeguards everywhere, a human able to check every decision. We moved the decision step from an LLM to Jev (@typesafeai): 25 typed questions, 0.29 s, $0.00015, code walks the workflow. Method, numbers, limits