BIRN actively is expanding its collaboration, authorization and user identification capabilities, which we expect to make available by early 2010. These new offerings will help us track, manage and search new and existing BIRN capabilities. We also will be able to set and customize research data access controls, improving the filtering mechanism for who is able to reach specific data.
In the field of knowledge engineering, we are developing new information-mapping capabilities, which we intend to demonstrate with a tool that can answer sophisticated queries. While the initial proof-of-concept will involve NIH grant abstracts, we believe the approach will be applicable to many other purposes.
Currently Under Development
Collaboration, Authorization, and User Identification
- Capability Registry – The BIRN Capability Registry will provide a formal, dual-component system for registering and tracking available capabilities. A capability definition registry will include descriptions and documentation, while a deployment registry will track availability of each service. These registries can be accessed via the BIRN website for casual browsing and queries or via web application interfaces for auto-configuring tools
Data Sharing and Integration
BIRN will add several new tools in early 2010 that user teams can deploy to enhance their data access and sharing services. As with other team-oriented BIRN services, our user teams will operate these services for their respective communities.
- Data Synchronization – This new capability will enable BIRN users to synchronize files that reside in both source and destination storage systems. The tool compares the contents of two data directories using a variety of comparator options. It then creates a list of files to be transferred from the source to the destination and performs that action, which it uses to synchronize content on the two systems.
- Group Policy Enforcement – The policy enforcement system, which runs on storage servers, will retrieve user group information from a BIRN-wide Group Policy Management service to determine whether to allow a user’s request for data access or storage. (Please see the Group Policy Management section above.)
- Clinical Data Extraction – This tool will provide a vendor-neutral mechanism for transmission of medical images, diagnoses and related data between health care providers. Clinical Data Extraction creates an abstract layer between data, meta-data and users for linking multiple DICOM storage service providers (Repositories) and meta catalog databases (Registries). By decoupling this mechanism from health-provider databases, networks and IT resources, we can avoid interfering with participating institutions’ mission-critical workflows – while extending image workflows and knowledge discovery to the research community
- Structured Queries Across Heterogeneous Data Sources – A new mediator tool will allow investigators to pose structured queries, using metadata terms, across multiple data repositories. The tool will negotiate the metadata syntax used in the query and repositories, then appropriately translate terms to retrieve relevant data from all specified repositories.
Knowledge-Based Services
BIRN developers are creating both general-purpose and application-specific Knowledge Engineering tools. Capabilities will include support for access to external knowledge resources, such as scientific literature, grant information (see ‘Topic Maps for Grant Abstracts’ below) and ontologies. We also are developing novel methods for building local scientific knowledge bases for end-users.
- Topic Maps for Grant Abstracts– This Knowledge Engineering web application (www.nihmaps.org) provides a “topic map” that enables all grant proposals funded by NIH in a particular year to be queried in several ways. The map shows the similarity of topics between grants, providing an intuitive, interactive clustered representation. Users may view grant patterns based on changing coloring schemes, overlays and other viewing options. Future developments will include creating subject-specific MEDLINE maps for specific scientific communities and comparing these maps with others derived from CRISP. For more on Knowledge Engineering, see the Knowledge Engineering WG profile.
Areas of Exploration
BIRN is particularly interested in working with current or prospective BIRN users to develop the following capabilities:
- Analysis and Workflows (Michael Wilde, University of Chicago; wilde at mcs dot anl dot gov)
- Authorization (Security WG)
- Collaboration (Infrastructure Team)
- Data Auditing and Provenance (Michael Wilde, University of Chicago)
- Data Sharing and Management (Data Management WG)
- Data Integration and Queries (Information Integration WG)
- Knowledge Engineering (Knowledge Engineering WG)
- Policy-based Replication (Data Management WG)
- Semantic Reasoning (Knowledge Engineering WG)
If you are interested in collaborating, please contact the person or Working Group listed after each subject.
