Data Integration Quality Check Tool (DIQCT)

Data Integration Quality Check Tool (DIQCT) evaluates and enforces data quality standards to support integration of multi-source cancer-imaging datasets for reliable AI-based disease outcome prediction.

Key Features:

• Data Quality Conceptual Model: Defines specific metrics to assess dataset quality across relevant dimensions.
• Detailed Data-Collection Protocol and Rule Set: Specifies data-collection protocols and rule sets to ensure consistency, homogeneity, and proper integration across multi-source datasets.
• Quality Verification and Corrective Action Suggestions: Performs checks against predefined quality requirements and provides actionable corrective measures to address identified deficiencies prior to research or clinical use.

Scientific Applications:

• Cancer-imaging repositories: Integrates and improves data quality in large-scale cancer-imaging repositories to support development and deployment of AI tools for disease outcome prediction.
• INCISIVE project: Applied within the INCISIVE pan-European project to support creation of a centralized cancer-imaging repository and to enhance data readiness for AI-driven outcome prediction.

Methodology:

Development of a Data Quality Conceptual Model with specific metrics; creation of data-collection protocols and rule sets to ensure homogeneity and proper integration; implementation of quality checks that verify compliance with established standards and suggest corrective actions.