> ## Documentation Index
> Fetch the complete documentation index at: https://developer.eka.care/llms.txt
> Use this file to discover all available pages before exploring further.

# SNOMED CT

## Overview

**SNOMED CT** (Systematized Nomenclature of Medicine — Clinical Terms) is the
most comprehensive clinical terminology in the world. Eka Care's codification
pipeline links free-text clinical entities — symptoms, findings, disorders and
procedures — to SNOMED CT concept identifiers.

| Property   | Value               |
| ---------- | ------------------- |
| `ontology` | `snomed`            |
| `version`  | `20250401_extended` |
| `metadata` | Not used            |

## When to use it

Use SNOMED CT to code clinical entities recorded during consultations:

* Patient-reported symptoms (`chest pain`, `shortness of breath`)
* Clinical findings and disorders
* Procedures and interventions

SNOMED CT is the right choice when you need a precise, granular clinical concept
rather than a billing-oriented code.

## How it works

The query is normalised, candidate concepts are retrieved through semantic
vector search over the SNOMED CT index, and a language model selects the single
best match. No `metadata` is required — send the clinical entity as the `query`.

## Reading the results

Each result is a SNOMED CT concept:

* `term_id` — the SNOMED CT concept identifier (e.g. `29857009`)
* `term_name` — a human-readable term for the concept
* `score` — semantic relevance of the candidate
* `is_linked` — `true` when the pipeline is confident this is the correct match
* `metadata.semantic_tag` — the concept's category (`finding`, `disorder`,
  `procedure`, …)

The pipeline also matches localised and transliterated terms, so a single
concept can be returned under multiple `term_name` variants across languages.

## Example

A query of `chest pain` links to SNOMED CT concept `29857009` ("Chest pain"),
tagged as a `finding`.

## Try it out

See the [Link Entity API](/api-reference/health-ai/medical-entity-codification/link-entity)
and select the **SNOMED CT** example to try it.