Medications
Goal of Extraction
The Core Clinical Medication Library is a medication-focused clinical data extraction tool designed to identify, extract, and ontologically map medications and pharmaceutical treatments from unstructured clinical narratives and patient health record documentation. It extracts medication names and dosing information documented across clinical notes, medication lists, and medication history sections, standardizing representation through RxNorm concept mapping.
Extraction Methodology
The extraction process operates at the sentence level, capturing each medication with precise character-level locations (start and end positions) that enable direct traceability back to the “document”. Each extracted medication record includes the medication name, dosing characteristics (dose, frequency, quantity, route, mode, duration), clinical indication, action, and temporal context (action date, date from note). A medication record will only be extracted if at least one of the following dosing attributes is present alongside the name of the drug: [dose, frequency, quantity, route, mode, duration]. When a patient’s medication appears multiple times within a single document—whether from repeated orders, serial administration, or documentation across different note sections—each instance is extracted and recorded separately to preserve the complete medication history. The library performs extraction on a document-by-document basis; there is no automatic cross-document reconciliation or cross-patient data merging.
Note: A document is determined by our intelligent document splitting feature. Medical PDFs can contain a single document or multiple documents. A document’s type can provide clinical context important for the data extraction. This feature separates and categorizes an input PDF into individual documents by document type to prepare them for data extraction. An input PDF can remain one document or can be split into multiple documents through this process. A “document” in the context of this data extraction refers to the documents identified after the intelligent document splitting process.
Interpretation & Validation
Every extracted medication record captures medication names and dosing characteristics. This extracted data guides appropriate reliance on medication history and supports informed clinical interpretation.
Clinical Considerations
Temporal context is critical for clinical decision-making; understanding when medications were administered or prescribed, along with duration information, directly impacts assessment of treatment regimen and medication adherence. Serial medication documentation provides essential context for understanding the medication history, treatment changes, and clinical response over time. Dosing characteristics (dose, frequency, route, mode) should be considered in conjunction with extracted indication to ensure appropriate clinical interpretation.
Target Concepts
Coding System: RxNorm
All extracted medications are mapped to RxNorm (National Library of Medicine’s normalized naming system for clinical drugs), with each medication assigned a valid RxNorm concept identifier, along with its associated full medication description and name. This structure provides both the standardized data needed for system integration and the contextual information needed for informed clinical interpretation.
Clinical Concepts Extracted
Medications are clinically documented pharmaceutical treatments that have been prescribed, administered, or documented. This includes medication names, dosing information, clinical indications, and medication actions. Medication names, dosing characteristics (dose, frequency, route, mode, duration), modifiers, and clinical indications are captured as they appear in source documentation. Medication action is standardized to controlled vocabulary.
We define medications with RxNorm concepts.
Extracted Fields
Medications Extraction Fields
Example of Extracted Output
- Note: This example demonstrates the content of the output, not the structure
Medications Illustrative Output Example
- Example JSONL object showing a typical extraction result
