caBIG® IN ACTION:
At the Kimmel Cancer Center at Thomas Jefferson University (TJU), the biospecimen collections are maintained in tissue banks dispersed across the various departments of the hospital, making it difficult for researchers to find the specific specimens they need to support their research. TJU turned to caTissue to provide a simple, userfriendly interface to manage and query their diverse biospecimen collections. Jack W. London, Ph.D., Director of the Informatics Shared Resource at the Kimmel Cancer Center, comments: "For us at Jefferson, caTissue Suite answered a need to provide a comprehensive tool for our researchers to identify specimens required for their research. We've been able to link three internal tissue banks already, and we are expanding our capabilities this year. The ability to share this information using caGrid is very valuable to our collaborations with other institutions."
caBIG® IN ACTION:
At the University of California at San Francisco, Max Wintermark, M.D., Director of the NeuroCardioVascular Imaging Laboratory, and the lead investigator working with caBIG®-enabled imaging tools, notes: "caBIG® is basically the response to what researchers like me have hoped for, for a long time. caBIG® can house multiple types of data, whether they are imaging or clinically oriented. Also, it is likely that in the future, other types of data will need to be stored (e.g., blood and genetic data), and caBIG® can accommodate these types of data. caBIG® enables data aggregated from multiple sites to appear as an integrated research tool set or large database while individual resources remain under the control of the originating organization. Thus, there could potentially be no significant upfront cost associated with a Center's decision to participate in a caBIG®-based research system."
The Jackson Laboratory Cancer Center studies mouse genetics to develop disease models that improve the prevention, diagnosis, and early detection of cancer in humans. The Jackson Laboratory has deployed caArray to make its internal data repositories available through caGrid. Chuck Donnelly, Director of Computational Sciences, discusses how caBIG® is transforming research at Jackson Labs: "The deployment of caArray has allowed us to create a state-of-the-art microarray analysis system for our research staff by providing a means of simplified access to data analysis tools that were previously unavailable. In addition to driving a shift in the way research is taking place at our lab, caArray has also enabled us to make our data available to other institutions. We believe this data will enable cancer research by informing pre-clinical and clinical studies in humans, essentially initiating a paradigm-shift in the way research is currently taking place."
Data Interoperability: Overcoming the Tower of Babel
Efficient and effective use of biomedical data is essential to understanding the molecular processes that underlie cancer and to developing better interventions. This ability to access and use data between different systems is termed "semantic interoperability." But multiple data representations and software products have evolved over the years, seriously limiting interoperability.
caBIG® enables data interoperability by systematically adopting internationally recognized data standards where they exist and by working closely with professional organizations to develop vocabularies, define common data elements (CDEs), and extend data models where such standards do not exist or are inadequate to the task. For example, in partnership with key industry standards organizations such as the Clinical Data Interchange Standards Consortium (CDISC) and Health Level 7 (HL7), caBIG® has helped create the Biomedical Research Integrated Domain Group (BRIDG) (http://bridgmodel.org/). caBIG® develops tools that adhere to these standards, allowing multiple institutions to exchange data and research collaboratively, similar to the way agreed-upon interoperable standards enabled the banking system to offer the interconnected network of ATMs that enable us to access our money anywhere in the world, regardless of an individual bank's location.
caBIG® provides two complementary pathways to achieving interoperability. Institutions may "adopt" caBIG® tools and use those applications to connect to caGrid, or they may "adapt" commercial or in-house developed software to be caBIG®-compatible. Many institutions pursue both approaches, depending on their specific needs and their existing IT infrastructure. A number of the NCI-designated Cancer Centers currently connecting to caGrid are working with commercial software vendors to ensure essential components of their IT infrastructure are fully-caBIG® compatible. The choice of pathway by a particular institution depends on many factors, including:
- The extent to which existing tools meet the needs of the end users or if additional capabilities are required.
- The receptiveness of end-user researchers to learn new tools and processes, or their desire to continue using familiar tools and workflows.
- The available resources for internal development or customization of existing tools to become caBIG®-compatible, versus installing existing caBIG® tools.
caBIG® has developed Application Programming Interfaces (APIs) and Software Development Kits (SDKs) that provide detailed instructions about application interfaces, simplifying the process of developing new caBIG®-compatible software, or modifying existing software to become caBIG®-compatible.
Recognizing that many organizations have a well-established IT infrastructure that satisfies some of their requirements, but still must provide interoperability within that existing infrastructure, the caBIG® program developed the caBIG® Integration Hub and caAdapter programs to address these needs. caBIG® Integration Hub generates messages in the industry standard HL7v3 format and simplifies the mapping and exchange of clinical trials information between applications and point-of-care systems. A Lab Viewer is also provided to assist in the inspection of clinical laboratory data. caAdapter is an open-source toolkit that integrates tightly with caBIG® Integration Hub and provides developers with web services and APIs to simplify application integration and model mapping services between different kinds of data sources.
3 Etheredge, Lynn M. 2009. Medicare's Future: Cancer Care. Health Affairs, 28(1):148-159.
