EHRchitect: An open-source software tool for medical event sequences data extraction from Electronic Health Records

Area of Science:

• Biomedical Informatics
• Health Data Science
• Computational Medicine

Background:

• Electronic Health Records (EHR) data is crucial for medical research but requires significant quality control and restructuring.
• Analyzing the sequence of medical events in EHR data is challenging yet vital for understanding disease progression and treatment outcomes.
• Current methods for EHR data mining and preparation are complex and time-consuming, hindering research efficiency.

Purpose of the Study:

• To introduce EHRchitect, a Python application designed to automate the transformation and quality control of exported EHR datasets.
• To facilitate the creation of optimized MySQL databases from EHR data for efficient analysis.
• To enable the extraction of sequential medical event data tailored to specific research configurations.

Main Methods:

• Developed EHRchitect as a Python application for automated EHR dataset transformation.
• Implemented functionality to create clean, formatted, and optimized MySQL databases.
• Enabled user-configurable sequential data extraction, including event sequences, patient metadata, and event metadata.

Main Results:

• EHRchitect successfully creates optimized databases for sequential medical event data extraction.
• Extracted data is provided in distributed Parquet files, containing comprehensive event and patient information.
• The application supports concurrent processing, allowing for effortless scaling on multi-processor systems.

Conclusions:

• EHRchitect significantly streamlines the processing of large EHR datasets for research.
• The tool offers a flexible framework for configuring event and timeline parameters for sequential data extraction.
• By automating quality control and simplifying data preparation, EHRchitect substantially reduces the time and effort required for EHR data analysis.