I have a question about parsing two specific variables from the Product Labeling Endpoint on OpenFDA:
- indications_and_usage
- dosage_and_administration
Ideally I would like to:
Extract each indication and its abbreviation. E.g. drug: Keytruda would be Melanoma, Non-Small Cell Lung Cancer (NSCLC), etc..
Extract each dosage (related to indication and population)
Map the indication with the dosage
The request via the API returns a long text/string, which is unstructured and difficult to parse.
Some questions I have:
Are there any existing libraries that can parse these two variables out?
Is there a paid service that OpenFDA offers that contains a parsed or more structured format of these two variables?
Are there any existing services by other companies that parses this information into a structured format?
I've already looked at the DailyMed XML files for these two sections (indications and dosages) and tried to parse it that way, however, the structure of the XML files for each drug is different and also difficult to parse.
Is this still the only way: Can we get the dosage list for a drug