Entity extraction from medical content using API calls

Job ID: 33668269

Budget: $15 – $25 USD

CLASSIC NLP ENTITY EXTRACTION & LABELLING USE CASE FOR HEALTHCARE NICHE

We want to extract and label entities within text. Data is provided in an excel column, containing html of input text (100-5000 words per unit). Total units of text - 1k to start with.

Use of Textrazor.com for extraction.

We expect output to be - Html text now modified with

1)Spell checked with automated replacements for suggestions with high confidence (have a couple of tools in mind but need suggestions, keeping in mind the medical nature of queries). Use of medical public dictionaries, if applicable. Options - https://prowritingaid.com/en/App/API https://www.perfecttense.com/api https://languagetool.org/proofreading-api
But we have to see if they are cost effective. Prowritingaid seems to be fine, if it works. Suggestions would be appreciated.
2)Grammar correction - ex using https://cloud.prowritingaid.com/analysis/swagger/ui/index

The above goal is structured and defined. But going ahead we need someone who has problem solving and coding chops to execute problems such as the following -

We have a library of keywords, medical terms, search terms which we want to label to this text. For example - “medications during pregnancy”. The text provided won’t contain this term specifically but would mention pregnancy and drugs so it needs to be labelled as such - preferably when text is found such as Adderall in pregnancy; it should be hyperlinked as medications in pregnancy. This could be done using entity extraction of Adderall as a drug (via Textrazor).


You may find an example task with detailed scope of the project below.
https://docs.google.com/document/d/1qQ0fgjUApPnH1OPsZBrO9Njn12PnAOjGFoHFhzuGXZ4/edit

Please give examples of any similar entity extraction work done by you in past.
Mention " I read about your entity extraction project" to stand out from spam bids.