发表机构
McGill University; Northwestern University; Ohio University（麦吉尔大学; 西北大学; 俄亥俄大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对生物医学语言模型未显式保留医学代码层级的问题，提出双曲临床本体嵌入（HCOE），将BioBERT嵌入映射至庞加莱球，结合本体引导对比学习与路径聚合，在ICD/ATC关系预测及MIMIC-IV多项临床任务上取得最优性能。
AI 中文摘要
生物医学语言模型（LMs）编码了文本语义，但并未显式保留医学代码的层级结构。我们提出了双曲临床本体嵌入（HCOE），用于层级感知的临床概念表示。HCOE将冻结的BioBERT嵌入映射到庞加莱球中，结合了父侧和子侧的本体引导对比学习以及从粗到细的本体路径聚合。它使用了由临床分类软件（CCS）组织的国际疾病分类（ICD）代码和解剖治疗化学（ATC）药物层级。评估表明，HCOE在ICD/ATC临床关系预测和CCS到PheCode层级迁移上表现最佳。在MIMIC-IV数据集上，HCOE在死亡率预测、再入院预测、药物推荐和罕见药物预测方面也取得了最佳性能。
英文摘要
Biomedical language models (LMs) encode textual semantics but do not explicitly preserve medical code hierarchies. We present Hyperbolic Clinical Ontology Embeddings (HCOE) for hierarchy-aware clinical concept representation. HCOE maps frozen BioBERT embeddings into a Poincare ball, combining parent-side and child-side ontology-guided contrastive learning with coarse-to-fine ontology-path aggregation. It uses International Classification of Diseases (ICD) codes organized by Clinical Classifications Software (CCS) and Anatomical Therapeutic Chemical (ATC) medication hierarchies. Evaluations show that HCOE performs best on ICD/ATC clinical relation prediction and CCS-to-PheCode hierarchy transfer. On the MIMIC-IV dataset, HCOE also achieves the best performance on mortality prediction, readmission prediction, medication recommendation, and rare drug prediction.
CommentsAccepted at IEEE BIBM 2026. 7 pages, 3 figures, 4 tables
