How to use Text Analytics for health

Important

Text Analytics for health is a capability provided “AS IS” and “WITH ALL FAULTS.” Text Analytics for health is not intended or made available for use as a medical device, clinical support, diagnostic tool, or other technology intended to be used in the diagnosis, cure, mitigation, treatment, or prevention of disease or other conditions, and no license or right is granted by Microsoft to use this capability for such purposes. This capability is not designed or intended to be implemented or deployed as a substitute for professional medical advice or healthcare opinion, diagnosis, treatment, or the clinical judgment of a healthcare professional, and should not be used as such. The customer is solely responsible for any use of Text Analytics for health. The customer must separately license any and all source vocabularies it intends to use under the terms set for that UMLS Metathesaurus License Agreement Appendix or any future equivalent link. The customer is responsible for ensuring compliance with those license terms, including any geographic or other applicable restrictions.

Text Analytics for health can be used to extract and label relevant medical information from unstructured texts, such as: doctor's notes, discharge summaries, clinical documents, and electronic health records. There are two ways to utilize this service:

• The web-based API and client libraries (asynchronous)
• A Docker container (synchronous)

Features

Text Analytics for health performs Named Entity Recognition (NER), relation extraction, entity negation and entity linking on English-language text to uncover insights in unstructured clinical and biomedical text. See the entity categories returned by Text Analytics for health for a full list of supported entities. For information on confidence scores, see the transparency note.

Tip

If you want to start using this feature, you can follow the quickstart article to get started. You can also make example requests using Language Studio without needing to write code.

Determine how to process the data (optional)

Specify the Text Analytics for health model

By default, Text Analytics for health will use the latest available AI model on your text. You can also configure your API requests to use a specific model version. The model you specify will be used to perform operations provided by the Text Analytics for health.

Supported Versions	latest version
2021-05-15	2021-05-15

Text Analytics for health container

The Text Analytics for health container uses separate model versioning than the REST API and client libraries. Only one model version is available per container image.

Endpoint	Container Image Tag	Model version
/entities/health	3.0.016230002-onprem-amd64 (latest)	2021-05-15
/entities/health	3.0.015370001-onprem-amd64	2021-03-01
/entities/health	1.1.013530001-amd64-preview	2020-09-03
/entities/health	1.1.013150001-amd64-preview	2020-07-24
/domains/health	1.1.012640001-amd64-preview	2020-05-08
/domains/health	1.1.012420001-amd64-preview	2020-05-08
/domains/health	1.1.012070001-amd64-preview	2020-04-16

Input languages

Currently the Text Analytics for health only supports the English language.

Submitting data

To send an API request, You will need your Language resource endpoint and key.

Note

You can find the key and endpoint for your Language resource on the Azure portal. They will be located on the resource's Key and endpoint page, under resource management.

Analysis is performed upon receipt of the request. If you send a request using the REST API or client library, the results will be returned asynchronously. If you're using the Docker container, they will be returned synchronously.

When using this feature asynchronously, the API results are available for 24 hours from the time the request was ingested, and is indicated in the response. After this time period, the results are purged and are no longer available for retrieval.

Getting results from the feature

Depending on your API request, and the data you submit to the Text Analytics for health, you will get:

Named Entity Recognition

Named Entity Recognition detects words and phrases mentioned in unstructured text that can be associated with one or more semantic types, such as diagnosis, medication name, symptom/sign, or age.

Relation Extraction

Relation extraction identifies meaningful connections between concepts mentioned in text. For example, a "time of condition" relation is found by associating a condition name with a time or between an abbreviation and the full description.

Entity Linking

Entity linking disambiguates distinct entities by associating named entities mentioned in text to concepts found in a predefined database of concepts including the Unified Medical Language System (UMLS). Medical concepts are also assigned preferred naming, as an additional form of normalization.

Text Analytics for health supports linking to the health and biomedical vocabularies found in the Unified Medical Language System (UMLS) Metathesaurus Knowledge Source.

Assertion Detection

The meaning of medical content is highly affected by modifiers, such as negative or conditional assertions which can have critical implications if misrepresented. Text Analytics for health supports three categories of assertion detection for entities in the text:

• Certainty
• Conditional
• Association

Service and data limits

For information on the size and number of requests you can send per minute and second, see the service limits article.

See also

• Text Analytics for health overview
• Text Analytics for health entity categories