News

MILA Seminar - Lena Baum on HERALD, a domain-specific query language for longitudinal health


  • Date: 10.03.2025
  • Time: 13:30-14:30
  • Speaker: Lena Baum
  • Title: HERALD: A domain-specific query language for longitudinal health

 

Analysis of longitudinal health data provides valuable insights into disease progression, treatment response, and prevention. However, transforming this data into cross-sectional tables for statistical analysis and machine learning remains complex. Existing tools often lack a balance between usability and functionality, making data extraction and transformation a challenge for researchers.
HERALD is a domain-specific query language designed to simplify this process. Its natural language-inspired syntax supports selection, aggregation, relationship analysis, filtering, and temporal constraints. Queries are executed individually for each patient, generating cross-sectional tables that can be used for further analysis. The system includes a web-based user interface for query construction, integrated statistical analysis, and integrates seamlessly with Informatics for Integrating Biology and the Bedside (i2b2).
As an open source tool, HERALD provides a practical solution for transforming longitudinal health data into cross-sectional tables, making complex data more accessible for medical research.