The Scope of In-Context Learning for the Extraction of Medical Temporal Constraints

Parker SeegmillerJoseph GattoMadhusudan BasakDiane CookHassan GhasemzadehJohn StankovicSarah Preum

Parker SeegmillerJoseph GattoMadhusudan Basak ...+3 Sarah Preum

Mar 2023

摘要

Medications often impose temporal constraints on everyday patient activity.Violations of such medical temporal constraints (MTCs) lead to a lack oftreatment adherence, in addition to poor health outcomes and increasedhealthcare expenses. These MTCs are found in drug usage guidelines (DUGs) inboth patient education materials and clinical texts. Computationallyrepresenting MTCs in DUGs will advance patient-centric healthcare applicationsby helping to define safe patient activity patterns. We define a novel taxonomyof MTCs found in DUGs and develop a novel context-free grammar (CFG) basedmodel to computationally represent MTCs from unstructured DUGs. Additionally,we release three new datasets with a combined total of N = 836 DUGs labeledwith normalized MTCs. We develop an in-context learning (ICL) solution forautomatically extracting and normalizing MTCs found in DUGs, achieving anaverage F1 score of 0.62 across all datasets. Finally, we rigorouslyinvestigate ICL model performance against a baseline model, across datasets andMTC types, and through in-depth error analysis.

展开全部

图表提取

论文十问

参考文献

被引用

提问与回答

暂无人提供速读十问回答

被引用

笔记

问答

领域