search

royashcenazi / parsigs

Parsigs is an open-source project that aims to extract the relevant dosage information from prescriptions text without compromising the patient׳s privacy.
MIT License
18 stars 10 forks source link

Brand Recognition - Data Collection. #22

Open itay-goldraich opened 10 months ago

itay-goldraich commented 10 months ago

From Known Issues

While Parsigs aims to identify and structure dosage instructions effectively, it's worth noting that the identification of brand names as part of the definition may not be as complete as the extraction of frequency, dosage, and period information (as this data is often not explicitly provided in the Sig, e.g., "Take 1 tablet every day").

I would like to have in the data folder a list of all/most/as big as you can make and find it to be of name-brand-medication. preferably a table or a list of drug name: drug company: generic name:

royashcenazi commented 10 months ago

@itay-goldraich in case we will have all of the drug names, It wont be easy to infer which drug is in the Sig how did you thought using this here?

itay-goldraich commented 10 months ago

It will have 2 main uses.

  1. The autocorrect feature isn't familiar with brand name medications and might make a mistake in confusioning a name with a word.
  2. Farmacist usually get a brand name drug from the doctor, but they can (and often do) change the drug to a generic or other drug with the same effect. This data can help with a future feature for such common action.