HEALTHCARE

Healthcare data is notoriously difficult to aggregate because of exclusive health record systems that store data differently. Not only that, each physician has their system for taking notes and shorthand notations that they're accustomed to.

​

That is why the vestibular database that we developed is so novel. We took thousands of patient's clinical visit notes and standardized the information to be aggregated into one database. Once standardized with a clinical template, I developed a Python program that reads in all of the files and parses out hundreds of variables. This database has been used for 8 national presentations, 4 published papers, and continues to be mined for novel analysis.

​

The program was developed in a way that allows for easy processing of updated data. Every half year or so, we update the database with new information such as follow-ups for existing patients or new vestibular patients. Additionally, new variables can be added to the database (including existing patients by matching on their unique identification number) to provide further analysis.