Overview
My objective was to utilize datasets from the MIMIC Hospital EHR to create a relational database within Microsoft SQL Server Management Studio. My goal was to automate the creation of the database and the loading of each table’s contents. This database will serve as a mock data warehouse for various future projects highlighting my ability to manage information through the complete life cycle of data.
Data & Tools
The Medical Information Mart for Intensive Care dataset was originally sourced from Hugging Face. I have not been able to find the original version I used. Similar versions can be found on the National Institute of Health or Physio.net. It is comprised of six parquet files that can be linked on unique identifiers. The database was created in Jupyter Notebook using Python and connected to Microsoft SQL Server Management Studio.
Methodology
The libraries used to create the MIMIC Hospital EHR Database include pandas, pyodbc and numpy. The datasets Triage Assessments, ED Stays, Radiology Notes, Diagnoses, Discharge Instructions, two different Discharge Summaries were cleaned and combined on unique, composite identifiers. Pyodbc is used to connect to SQL Server Management Studio, create the database and each individual table. It is then used to insert column values for each table respectively. The connection is closed once complete.
Code Preview
Full Code & Diagram
*The code is around 80 pages due to the unstructured clinical notes.
