Skip to main content

Overview

This endpoint finds medical concepts that are semantically similar to a specific concept identified by its OMOP concept ID. It uses the same advanced machine learning algorithms as the POST version but optimized for concept-to-concept similarity matching, making it ideal for discovering related concepts when you have a specific starting point.

Path Parameters

conceptId
integer
required
The OMOP concept ID to find similar concepts for

Query Parameters

vocabulary_ids
string
Target vocabularies to search within (comma-separated)
Examples: SNOMED, SNOMED,ICD10CM, RXNORM,NDC
domain_ids
string
Clinical domains to focus the similarity search (comma-separated)
Examples: Condition,Procedure, Drug,Device
concept_class_ids
string
Concept classes to include in similarity search (comma-separated)
Examples: Clinical Finding,Procedure, Ingredient,Brand Name
similarity_threshold
number
default:"0.7"
Minimum similarity score threshold (0.0 to 1.0)
Higher values = More strict similarity matching
page_size
integer
default:"20"
Number of similar concepts to return per page
page
integer
default:"1"
Page number (1-based indexing)
include_scores
boolean
default:"true"
Include similarity scores in the response
include_explanations
boolean
default:"false"
Include explanations for why concepts are considered similar
standard_concept
string
Filter to standard concepts only
Options: S (standard), C (classification), NULL (non-standard)
include_invalid
boolean
default:"true"
Include invalid/deprecated concepts in similarity search
algorithm
string
default:"hybrid"
Similarity algorithm to use
Options: semantic (embedding-based), lexical (text-based), hybrid (combined)
exclude_self
boolean
default:"true"
Exclude the source concept from results

Response

success
boolean
Indicates if the request was successful
error
object
Error information (present only on error responses)
data
object
Contains the similar concepts search results

Usage Examples

Basic Similarity by ID

Find concepts similar to Type 2 Diabetes (concept ID 44054006):

Cross-Vocabulary Similarity

Find similar concepts across multiple vocabularies:
Find similar pharmaceutical concepts:

High-Precision Similarity

Get only highly similar concepts with detailed scoring:

Performance Notes

  • Concept-to-concept similarity is typically faster than query-based similarity
  • Hybrid algorithm provides best balance of accuracy and performance
  • Semantic algorithm is most accurate but computationally intensive
  • Caching is applied for frequently requested concept similarities