Skip to main content

Medical Terminology Server

Every healthcare organization develops its own vocabulary: local drug codes, institution-specific procedure names, abbreviated diagnoses, and custom billing codes. A cardiologist might document "MI", "AMI", "heart attack", or "myocardial infarction" to describe the same condition. A pharmacy system might use "ACEI-10" while the EHR lists "lisinopril 10mg tabs" for the identical medication.

Without a common language, these variations become barriers to analytics, research, and interoperability. The Medical Terminology Server solves this by normalizing all clinical concepts: whether extracted from free text or structured fields: to standard medical vocabularies, ensuring consistent representation across your entire organization and enabling seamless data exchange with external partners.

Why Terminology Normalization Matters​

Consider a cohort query for "patients with diabetes". Without terminology normalization, you'd need to search for:

  • Diabetes mellitus
  • Type 2 diabetes
  • T2DM
  • NIDDM
  • DM2
  • Adult-onset diabetes
  • Non-insulin dependent diabetes

And dozens more variants documented across clinical notes, problem lists, and billing systems. Miss even a few, and your cohort is incomplete.

With terminology normalization, all these variants automatically map to SNOMED CT concept 44054006 (Diabetes mellitus type 2). Query once, capture everything. This same principle applies to medications, lab tests, procedures, and every other clinical concept in your data.

Supported Standard Vocabularies​

The platform normalizes clinical concepts to five core medical terminologies, each optimized for specific clinical domains:

SNOMED CT

Comprehensive clinical terminology for conditions, findings, procedures, and anatomical structures.

Example: "Congestive heart failure" → SNOMED 42343007

RxNorm

Standardized drug names, ingredients, and medication products.

Example: "Lisinopril 10mg tablets" → RxNorm 314076

LOINC

Universal codes for laboratory tests, clinical measurements, and observations.

Example: "Hemoglobin A1c" → LOINC 4548-4

ICD-10-CM

Diagnosis codes for regulatory reporting and billing interoperability.

Example: "Type 2 diabetes with neuropathy" → ICD-10 E11.40

CPT

Procedure codes for billing, quality measurement, and outcomes tracking.

Example: "Coronary artery bypass graft" → CPT 33533

40+ Additional Vocabularies

Extended support for specialized terminologies including NDC, HCPCS, MedDRA, HPO, UMLS, NCI Thesaurus, MeSH, ICD-O-3, ICD-10-PCS, ATC, CVX, and more.

Map Clinical Concepts to Standard Vocabularies​

The terminology server operates seamlessly in the background, automatically normalizing concepts as they're extracted or transformed:

1

Extract or Receive Clinical Concept

NLP extracts "metformin 1000mg BID" from clinical note, or structured field contains local code "MET-1000"

↓

2

Search Terminology Server

Query across vocabularies to find matching standard concepts based on name, synonym, or code

↓

3

Map to Standard Code

RxNorm 861007: Metformin 1000 MG Oral Tablet

↓

4

Store in OMOP CDM

Concept persisted with standard vocabulary ID, enabling consistent querying across all sources

Key Capabilities​

Search across all vocabularies by name, code, synonym, or description. The server handles spelling variations, abbreviations, and common clinical shorthand automatically.

Example queries:
  • Search for "hypertension" → Returns SNOMED concepts for essential hypertension, secondary hypertension, pulmonary hypertension, and related conditions
  • Search for "A1c" → Returns LOINC codes for hemoglobin A1c tests (whole blood, serum, multiple methodologies)
  • Search for "ace inhibitor" → Returns RxNorm drug classes and individual ACE inhibitor medications

The search engine understands clinical context and returns ranked results based on relevance and semantic similarity.

Concept Mapping​

Map local institutional codes to standard concepts, preserving institutional vocabulary while enabling standardized analytics.

Mapping examples:

Local pharmacy codes:

  • "ACEI-10mg" → RxNorm 314076 (Lisinopril 10 MG Oral Tablet)
  • "STATIN-HIGH" → RxNorm 859424 (Atorvastatin 80 MG Oral Tablet)

Local diagnosis codes:

  • "HTN-benign" → SNOMED 59621000 (Essential hypertension)
  • "CHF-systolic" → SNOMED 441530006 (Chronic systolic heart failure)

Local lab codes:

  • "HGB-A1C" → LOINC 4548-4 (Hemoglobin A1c/Hemoglobin.total in Blood)
  • "TROP-I" → LOINC 10839-9 (Troponin I.cardiac [Mass/volume] in Serum or Plasma)

Once mapped, queries using standard concepts automatically include data coded with local institutional codes: no separate translation layer required.

Concept Relationships​

Navigate hierarchical and semantic relationships between concepts, enabling sophisticated queries and automated reasoning.

Relationship types:

Hierarchical (IS-A):

  • Metformin IS-A Biguanide IS-A Antidiabetic agent
  • Query for "antidiabetic agents" automatically includes all metformin products

Synonym (Maps to):

  • SNOMED 44054006 (Diabetes mellitus type 2) Maps to ICD-10 E11.9 (Type 2 diabetes without complications)
  • Enables cross-vocabulary queries and regulatory reporting

This relational structure powers advanced queries like "patients on any ACE inhibitor or ARB" without manually listing every drug product.

Vocabulary Versions​

The terminology server maintains current, validated versions of all standard vocabularies:

  • SNOMED CT: US Edition, latest release (updated annually)
  • RxNorm: Monthly updates from NLM
  • LOINC: Biannual updates (January and July releases)
  • ICD-10-CM: Annual updates (October 1 effective dates)
  • CPT: Annual updates from AMA

Version metadata is tracked for every concept, ensuring reproducibility and audit compliance. When vocabularies update, the system maintains backward compatibility while offering migration paths to new codes.

Automatic Integration​

The terminology server operates transparently across the entire platform. You don't explicitly call it: it's embedded in every data transformation:

Information Extraction

NLP-extracted entities are automatically mapped to SNOMED CT, RxNorm, and LOINC as they're extracted from clinical text.

Medical Reasoning

Inference engine uses concept relationships to identify drug classes, disease hierarchies, and related conditions for reasoning tasks.

OMOP CDM Transformation

All clinical facts stored in OMOP CDM reference standard concept IDs, ensuring consistent representation across the entire knowledge base.

Benefits of Terminology Standardization​

Without terminology normalization:

  • Incomplete cohorts: Queries miss patients whose data uses different terms for the same concept
  • Analytics fragmentation: Each project invents its own mapping logic, producing inconsistent results
  • Interoperability failure: Data can't be shared with external partners or registries
  • Research bias: Systematic undercounting of conditions documented with non-standard terms

With the terminology server:

  • Query once, capture everything: All variants automatically resolve to standard concepts
  • Consistent analytics: One shared vocabulary foundation across all secondary use applications
  • Regulatory readiness: Data already mapped to ICD-10, CPT, and other required code sets
  • Future-proof: New terminologies and updates integrate without breaking existing queries
Last updated on