Providing traceability for neuroimaging analyses

Area of Science:

• Neuroscience
• Medical Informatics
• Biomedical Research

Background:

• Increasingly digital biomedical data necessitates accurate information capture, traceability, and accessibility for researchers.
• Grid- and Cloud-based technologies, particularly Service Oriented Architectures (SOA), offer solutions for managing distributed data and algorithms in biomedicine.
• Neuroscientific image analysis requires robust traceability of processes and datasets for collaborative study, a capability often lacking.

Purpose of the Study:

• To address the need for deployed medical Grids that provide essential research data traceability for complex analyses.
• To outline the findings of a requirements study and a resulting system architecture for services supporting neuroscientific studies of Alzheimer's disease biomarkers.
• To develop and evaluate data management and provenance services for 21st-century medical research.

Main Methods:

• Developed a software infrastructure and services based on a Service Oriented Architecture (SOA).
• Integrated the CRISTAL software for provenance management within the SOA.
• Deployed the system to manage neuroimaging projects focused on Alzheimer's disease biomarkers.

Main Results:

• A Provenance Service was delivered within the neuGRID and N4U projects.
• The service captures and reconstructs workflow information essential for neuroimaging analyses.
• The system enables tracking of workflow and dataset evolution, as well as analysis outcomes.

Conclusions:

• The Provenance Service provides essential traceability throughout the lifecycle of neuroimaging studies.
• The generic design allows application across the broader medical domain as a reusable tool.
• Enables distributed collaborative medical analysis programs for research communities.