The Sublanguage of Clinical Problem Lists: A Corpus Analysis

Area of Science:

• Medical Informatics
• Natural Language Processing
• Clinical Linguistics

Background:

• Summary-level clinical text, or problem lists, offers a condensed view of patient health.
• These lists are crucial for understanding patient history and current condition.
• Identifying the specific language used in problem lists is key to unlocking their full potential.

Purpose of the Study:

• To explore the grammatical and semantic structure of clinical problem list entries.
• To identify the specialized rules, or "sublanguage," governing these entries.
• To develop methods for dynamically learning sublanguage features within specific medical concepts.

Main Methods:

• Large-scale corpus analysis of clinical problem list entries.
• Utilizing Resource Description Framework (RDF) for data representation.
• Incorporating inferencing and reasoning with domain-specific ontologies to find semantic patterns.

Main Results:

• Identified common semantic patterns within clinical problem list entries.
• Demonstrated the application of RDF and ontologies for sublanguage analysis.
• Explored dynamic methods for learning concept-specific sublanguage features.

Conclusions:

• Clinical problem lists possess a distinct sublanguage that can be analyzed using computational methods.
• Ontology-driven corpus analysis provides insights into the structure and semantics of clinical notes.
• These findings can enhance the extraction and utilization of patient information from electronic health records.