A web solution to harmonise different Alzheimer's Disease cohorts into a standard data schema, using ETL processes in a multi-institution environment.

GitHub Code here.

About BIcenter-AD

BIcenter is a web ETL tool using Pentaho Kettle as the DI execution engine. This tools offers:

  • Collaborative Environment: Develop ETL pipelines as a team.
  • Data Protection and Privacy: Data Sources and DI execution servers are decoupled from BIcenter.
  • RBAC: only allowed users have access to data pipelines and reports.
  • Centralized Installation: for system admins.
  • Easy to extend: Reflection-based approach to quickly embed new steps into the platform.

BIcenter-AD is a adpated version containing new components to semi-automatically harmonise large amounts of medical concepts in clinical studis.

It creates new opportunities for the study of rare conditions, where typically isolated cohorts do not provide enough statistical evidence.

The results can augment clinical knowledge by automatically computing new patient information during the migration stage.

BIcenter-AD can migrate and harmonize clinical cohorts from CSV format into the OHDSI OMOP CDM schema. This procedure increases the interoperability of the data by allowing the exportation of several cohorts into a new system reusing the same scripts.


6,669 subjects

398 cohort attributes

172 standardized concepts

Documentation

BIcenter-AD was created using BIcenter in its core. The documentation about the BIcenter's features, and guidelines for developers is available here.

Core Team

The methodology appied in BIcenter-AD was validated by the cohort data owners and developed by the following team members:

João R. Almeida

Researcher

José L. Oliveira

Full professor