Anchoring Clinical Events in Time: UID-Preserving Multimodal Reconstruction and Source-Grounded Adjudication· 临床事件时间锚定的多模态框架
Clinical timelines support treatment-window analysis and leakage-free modeling, but discharge summaries often obscure chronology and structured EHR tables describe only part of the patient course. We present a UID-preserving framework that links each narrative event occurrence to its source span and retains that identity through text-only estimation, structured-evidence retrieval, timestamped source-row grounding, and joint revision. We also present GAVEL, an LLM judge that compares two UID-aligned timelines against the narrative and structured record, to augment prior matching and temporal assessments. Across six open-weight models and 40 mixed-critical-care summaries, the GLM 5.2 multimodal revision, as compared to its text-only variant, improved temporal agreement without reducing event
提出保留UID的多模态框架和LLM判决器GAVEL，改进临床事件时间锚定。
- 核心方法
- 构建了一个保留UID的框架，能够将每个叙述性事件与源文本范围链接，并通过纯文本估计、结构化证据检索、时间戳源行锚定和联合修订保持一致性。引入GAVEL，一种LLM判决器，用于对比两个UID对齐的时间线与叙述和结构化记录，增强匹配和时间评估。
- 适合谁读
- 研究者、工程师
- 要解决的问题
- 临床事件的时间信息在出院摘要中常被模糊，且结构化EHR表格只能描述患者部分病程，影响治疗窗口分析和无泄漏建模。
- 关键实验
- 在6个开放权重模型和40个混合重症监护出院摘要上进行了实验，验证了GLM 5.2多模态修订方法相较于纯文本方法在时间一致性的提升。
- 主要贡献
- 开发了一种多模态方法，相较于纯文本方法，提高了临床事件的时间一致性和准确性，同时不减少事件的完整性。
- 意义与局限
- 该方法有助于更准确地分析临床时间线，支持更精细的治疗窗口分析和无泄漏建模。但需要进一步验证在更大规模数据集上的表现。