Overview

This guide introduces the Federated Data Quality Framework (FDQF) and its tooling.
It is intended for users who want to deploy, operate, or understand the system at a high level.

For deeper technical information, configuration guides, and deployment steps, please refer to the additional sections in the documentation menu.

---

What is the FDQF?

The Federated Data Quality Framework (FDQF) provides a structured approach and supporting software to assess and improve data quality in federated environments — situations where data remains distributed across independent locations due to legal, organizational, or technical constraints.

Instead of moving data to a central location, the FDQF enables local data quality analysis and privacy-preserving reporting across multiple sites.

Key Components

The FDQF consists of two main software components:

Data Quality Agent (DQA)

Runs locally at each data-holding site.

Responsibilities:

• Executes predefined, domain-specific data quality checks
• Converts human-readable rules into machine-readable queries
  (currently supported: HL7 CQL — Clinical Quality Language)
• Queries the connected database locally
• Generates aggregated and privacy-preserving results
• Pushes data quality reports to the central server (one-way; no data pull)

Privacy-First Approach

If you would like to learn how do we guarantee privacy preservation/anonymization of the shared results see the Privacy page.

Data Quality Server (DQS)

Runs centrally to collect and present results.

Responsibilities:

• Receives quality reports from multiple sites
• Aggregates results across the network
• Provides dashboards and views for end-users
• Provides a REST API for including the reports in external Dataset catalogues
• Enables researchers or study investigators to evaluate multi-site data quality without accessing raw data

Why Federated Data Quality?

Traditional data quality validation typically requires central access to raw datasets.
In many real-world environments — such as healthcare or regulated research settings — this is not possible.

The FDQF enables:

• Local processing at each data site
• Privacy-preserving quality metrics
• Cross-site comparison without data sharing
• Standardized and reproducible quality evaluation
• Support for federated biomedical and research infrastructures

Who Is This For?

The FDQF is designed for:

• Research networks and data consortia
• Healthcare institutions
• Federated biobanking and cohort infrastructures
• Privacy-sensitive or regulated environments
• Projects requiring data quality transparency without data transfer

Roles that may use the system include:

• Data Stewards
• Study Principal Investigators
• Clinical Researchers
• Data Engineers and IT Operators

In Summary

Concept	Description
Federated approach	Data stays at each site — processing happens locally
Local Agent	Executes checks and generates privacy-preserving metrics
Central Server	Collects, aggregates, and displays results
Goal	Enable cross-site data quality assessment without raw data sharing

The FDQF ensures trust, privacy, and transparency in multi-institution data ecosystems by combining local computation with centralized insight.

See for yourself

If you would like to experiment with the tooling please go to the Getting Started page.