How to Extract PubMed Articles and Apply Mesh Terms

About Mesh Tagger

• Mesh Tagger is used to return PubMed articles, classify, and return, and apply Mesh Term tags and similar articles. 
• Mesh Tagger calls an API from PubMed. For more information see https://id.nlm.nih.gov/mesh/swagger/ui. 
• MeSH (Medical Subject Headings) is the NLM-controlled vocabulary thesaurus used for indexing articles for PubMed
• To learn more about MeSH, also see https://www.nlm.nih.gov/bsd/disted/meshtutorial/introduction/index.html.
• The API for Mesh information and Similar Documents on PubMed uses: https://meshb.nlm.nih.gov/MeSHonDemand

Install the Mesh Tagger

How to Add the Mesh Tagger Component to a Pipeline

1. To add the Mesh Tagger component to a pipeline in AutoClassifier, see How to Add Components to Pipelines.
2. Select Mesh Tagger as your component when you are instructed to add your component in the section. 
3. Click Add Component.
4. The newly added component will appear in the pipeline components list

Mesh Tagger Pipeline Configuration

To configure your Mesh Tagger component, select it from the components list and complete the following fields in the Configuration section:

1. Api Endpoint: Specify the API endpoint. This is public.
2. PubMed Article Url Format: Enter a template you can use for the URL. You can also use this format to access the document in PubMed.
3. Input Property: Enter the name of the metadata to be used for processing. For example, "body".
4. Max Character Length: Enter the maximum number of characters used as text for the API request. For example, 5000. Note that the Mesh API is slower for a larger number of characters. Upland BA Insight recommends that this number does not exceed 5000. The maximum value allowed by the API is 10000
5. Send raw response as metadata: Enable this field to output from Python API in JSON format. 
6. Click Save.

Output Properties

Property	Type	Description
PubMedSimilarArticles	Text – Multi	Includes - from PubMed- multiple: Article IDs URLs Article Titles
PubMedMeshTerms	Text – Multi	Example: 'Human', 'Glucose tolerance test', 'Weight Loss'