5

I have a project with a police department in Washington State that is about to start where I will be automatically publishing police reports not requiring redaction per the state public records act. I have a simple filter, see https://gist.github.com/policevideorequests/6f6662b7ae673934cf0f. The only thing left is to detect medical words. Does anyone know of a simple public list of medical words including drug names?

  • 1
    Could you tell us more about what state you are looking for and what the law dictates? It isn't necessary for us to help you with information. But it would be an excellent anecdote for how someone proposes to use open data in the real world and would enrich our pool of knowledge. Also, could I ask what types of medicine and health related events you would expect to redact? – Kotebiya Feb 9 '15 at 16:46
2

The NLM's UMLS Metathesaurus http://www.nlm.nih.gov/research/umls/knowledge_sources/metathesaurus/ is a possibility but, there is some non-trivial technical overhead to using it. You can, however, get it at no cost. You might check out FDA's NDC database file. Perhaps that will help you. http://www.fda.gov/Drugs/InformationOnDrugs/ucm142438.htm You can also find a small list of drug ingredient names in the documentation for prescription medications collected by the National Health and Nutrition Examination Survey. As eluded to above, your request is quite broad so it is difficult to give you more concrete options.

3

First thought: ICD-10, the current international classification of diseases by WHO:

http://www.who.int/classifications/icd/en/

I am not sure if this helps, escpecially in regard to drug names.

3

drug names....technical terms or street terms? technical terms i'm assuming will be on cdc or openfda or perhaps both. i've been hearing about openfda much more lately, so i would check them out first.
here's a medical glossary that you could scrape, it seems pretty in-depth but i have no idea what i'm looking at and/or how or what to compare it to:
http://www.online-medical-dictionary.org/

3

This is a difficult question (one that is facing healthcare in the US right now -- taking free text and creating structured medical data suitable for analysis and reporting). There are many ontologies you can query against depending on your needs. It is a big endeavor regardless of how far you dive in. If you want a really formal, complete system (total overkill), look at ctakes,

http://ctakes.apache.org/

Otherwise, there are individual ontologies/metadata. For example, tools like RxNorm can take prescription information and map it to a standard. SNOmed is going to have the information for medical/anatomy/clinical terms

http://en.wikipedia.org/wiki/Systematized_Nomenclature_of_Medicine

And ICD, CPT other codes will be involved in billing standardization.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.