Hierarchical Annotation for Building A Suite of Clinical Natural Language Processing Tasks: Progress Note Understanding
This work addresses the problem of limited resources for training NLP models in clinical diagnostic thinking for researchers and practitioners in healthcare AI, though it is incremental as it builds on existing EHR annotation efforts.
The authors tackled the lack of annotated corpora for modeling clinical diagnostic thinking in EHR data by introducing a hierarchical annotation schema with three stages for clinical text understanding, reasoning, and summarization, resulting in a new annotated corpus based on publicly available daily progress notes and a suite of tasks called Progress Note Understanding.
Applying methods in natural language processing on electronic health records (EHR) data is a growing field. Existing corpus and annotation focus on modeling textual features and relation prediction. However, there is a paucity of annotated corpus built to model clinical diagnostic thinking, a process involving text understanding, domain knowledge abstraction and reasoning. This work introduces a hierarchical annotation schema with three stages to address clinical text understanding, clinical reasoning, and summarization. We created an annotated corpus based on an extensive collection of publicly available daily progress notes, a type of EHR documentation that is collected in time series in a problem-oriented format. The conventional format for a progress note follows a Subjective, Objective, Assessment and Plan heading (SOAP). We also define a new suite of tasks, Progress Note Understanding, with three tasks utilizing the three annotation stages. The novel suite of tasks was designed to train and evaluate future NLP models for clinical text understanding, clinical knowledge representation, inference, and summarization.