Federating distributed and heterogeneous information sources in neuroimaging: the NeuroBase Project.

Area of Science:

• Neuroscience
• Computer Science
• Information Science

Background:

• Neuroimaging data is often distributed across multiple research centers, leading to data silos.
• Heterogeneous data formats and processing programs hinder collaborative research in cognitive neurosciences.
• A unified approach is needed to integrate and access these distributed resources effectively.

Purpose of the Study:

• To define requirements for federating neuroimaging information sources via the internet.
• To develop a shared ontology for supporting diverse neuroimaging applications.
• To design a computer architecture for accessing and sharing distributed neuroimaging data.

Main Methods:

• Developed a shared semantic model and ontology for neuroimaging data.
• Designed a computer architecture utilizing a mediator/wrapper approach.
• Created a demonstrator to compare distributed image processing tools on distributed data.

Main Results:

• A functional architecture for federating distributed neuroimaging information.
• A shared ontology enabling standardized data representation and access.
• Demonstrated the feasibility of comparing distributed image processing tools.

Conclusions:

• The NeuroBase project provides a framework for integrating distributed neuroimaging resources.
• The proposed architecture facilitates collaborative research by enabling data and tool sharing.
• Future deployment will enhance neuroimaging research through improved data accessibility and analysis.