> ## 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.

# ICD-10-CM

## Overview

**ICD-10-CM** (International Classification of Diseases, 10th Revision, Clinical
Modification) is the standard code set for diagnoses. Eka Care's codification
pipeline links free-text diagnoses to ICD-10-CM codes.

| Property   | Value        |
| ---------- | ------------ |
| `ontology` | `icd-10-cm`  |
| `version`  | `comprehend` |
| `metadata` | Not used     |

## The Comprehend version

The `comprehend` version is built on **AWS Comprehend Medical**, a clinical NLP
service. Unlike a plain term lookup, it understands diagnoses expressed in
natural clinical language — handling modifiers, qualifiers and surrounding
context — which makes it well suited to coding from real clinical notes.

## When to use it

Use it to code diagnoses from free-text clinical content:

* Diagnoses recorded in consultation or discharge notes
* Problem lists and assessments
* Claims and medical-coding workflows

Send the diagnosis as the `query` — no `metadata` is needed.

## Reading the results

Each result is an ICD-10-CM code:

* `term_id` — the ICD-10-CM code (e.g. `E11`)
* `term_name` — the diagnosis description
* `score` — confidence of the candidate
* `is_linked` — `true` when the pipeline is confident of the match

The response also includes a `query_breakdown` derived from AWS Comprehend
Medical:

* `entity` — the clinical entity detected in the query
* `entity_score` — confidence of that detection
* `traits` — detected traits, e.g. `DIAGNOSIS`
* `attributes` — detected attributes, e.g. a `QUALITY` of `type 2`

## Example

A query of `type 2 diabetes mellitus` links to ICD-10-CM `E11`
("Type 2 diabetes mellitus"), with `query_breakdown` reporting the detected
entity `diabetes mellitus` and the attribute `type 2`.

## Try it out

See the [Link Entity API](/api-reference/health-ai/medical-entity-codification/link-entity)
and select the **ICD-10-CM Comprehend** example to try it.