AMIA2014 Workflow Poster (Tuesday) Automating Extraction and Calculation of Daily Dose and Duration for Medications in EHRs

(Quotes from 2014 AMIA proceedings that interest me due to workflow-related implications: Take me to my Workflow-Related #AMIA2014 Papers and Posters rationale!)

5:00 Tuesday

[CW: open source workflow]

“As part of the eMERGE (electronic Medical Records and Genomics) project, we are engaged in studies that require the extraction of medication data from EHRs (Electronic Health Records). In order to calculate daily dose and duration of medications, we found it necessary to supplement the existing NLP (Natural Language Processing) tool we are using with additional techniques. The result is an open-source workflow that can be readily re-targeted to multiple medications and is 90% accurate.

….

We created an executable workflow in KNIME (Konstanz Information Miner, knime.org) for calculating average daily dose and duration for medications, which extracts the numerical data needed for the calculations in 3 major steps: 1) combining the results of the MedEx NLP tool on prescription order signature line text with discrete medication data from our EHR where available, both of which are mostly non-numeric; 2) using pattern matching (RegEx), Java, and other data transformation tasks in KNIME, to convert dose and duration into numeric fields in standardized units; and 3) applying logical rules, for duration in particular, prioritizing fields to determine an accurate daily dose and duration.”

American Medical Informatics Association 2014 Proceedings

This entry was posted in amia2014. Bookmark the permalink. Post a comment or leave a trackback: Trackback URL.

Post a Comment

Your email is never published nor shared. Required fields are marked *

*
*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

You can add images to your comment by clicking here.