Teaching LLMs to Recommend and Defer in Underrepresented Epilepsy Care

  • 类型：arxiv
  • 标识：2606.31036
  • 链接：https://arxiv.org/abs/2606.31036
  • 主分类：agent
  • 形态：application
  • 被引：0
  • 被引来源：Semantic Scholar + OpenAlex
  • S2被引：0
  • OpenAlex被引：0
  • 影响力被引：0
  • TLDR：Specialist epilepsy expertise is scarce in resource-constrained settings, making LLM-based decision support attractive for frontline clinicians managing longitudinal treatment. Such systems must adapt to local prescribing practice and know when to defer. We study this problem in Ugandan pediatric epilepsy care, predicting anti-seizure medication regimens from longitudinal unstructured clinic notes. Standard prompting achieves non-trivial agreement with physician prescriptions, but neurologist review shows that many errors reflect distribution-miscalibrated prescribing defaults rather than fail
  • OpenAlex ID：W7166845450
  • OpenAlex DOI：10.48550/arxiv.2606.31036
  • DOI：10.48550/arxiv.2606.31036
  • DOI来源：OpenAlex
  • 开放获取：green
  • 开放获取链接：https://doi.org/10.48550/arxiv.2606.31036
  • OpenAlex更新：2026-07-19
  • 待LLM分类：否
  • 成熟度：research
  • 场景：clinical decision support、epilepsy care、LLM deferral
  • 标题中文：教导 LLM 在欠发达的癫痫诊疗中进行推荐与转诊
  • TLDR中文：在资源受限环境中，专业癫痫专家稀缺，使基于 LLM 的决策支持对管理纵向治疗的一线临床医生具有吸引力。此类系统必须适应当地处方实践并知道何时转诊。我们在乌干达儿科癫痫诊疗中研究该问题，基于纵向非结构化门诊记录预测抗癫痫用药方案。标准提示与医生处方取得了一定程度的一致性，但神经科医生审查显示许多错误反映的是分布失校的处方默认值而非失败。
  • 来源文件：
  • /inbox/tom/_candidates/2026-07-07-rag-retrieval-reranking-candidates.json
  • [S2 enrich]
  • [OpenAlex backfill]