Page History

...

The growing adoption of HL7 standards (for example, the HITSP/C32 specification called out in the meaningful use final rule) throughout the world suggests that aligning Semantic Infrastructure 2.0 around these specifications will streamline attainment of Semantic Infrastructure 2.0 objectives.

CDISC

CDISC is a global, open, multidisciplinary, non-profit organization that has established standards to support the acquisition, exchange, submission and archive of clinical research data and metadata. The CDISC mission is to develop and support global, platform-independent data standards that enable information system interoperability to improve medical research and related areas of healthcare. CDISC standards are vendor-neutral, platform-independent and freely available via the CDISC website.

CDISC defines standards to be used or reused in the definition, development, and execution of clinical trials as well as the submission of the clinical trial data to regulatory agencies (e.g., FDA) for drug, device or procedure approval.  CDISC is currently focused on constructing a Shared Health and Research Electronic Library
(SHARE) and would like to use the NCI Knowledge Repository as the basis for SHARE.

The CDISC SHARE project is attempting to map CDISC standards (e.g., SDTM & CDASH) to BRIDG concepts and then have these concepts reused by CDISC participants (e.g., CDOs, Pharmas), who create their own local concepts. BRIDG already incorporates a number of CDISC standards (e.g., STDM) and is working to harmonize other CDISC standards.

CDISC SHARE will contain the existing CDISC standards and will provide machine readable element (variables) within those standards. This will allow a range of applications used within other organizations (e.g., Clinical Research Organizations (CROs), Pharma, and other Agencies) to automatically access those definitions.

ISO 21090 Harmonized Data Types for Information Exchange

The NCI CBIIT Enterprise Conformance and Compliance Framework (ECCF) document requires that, in order to achieve Computable Semantic Interoperability (CSI), each attribute of a static information model should be bound to a robust data type specification.

The ISO 21090 standard provide a harmonized set of data type definitions used for representing and exchanging basic concepts that are commonly encountered in support of information exchange in healthcare environment. They represent a culmination of a large scale joint effort among the standard bodies such as HL7 and ISO, and have been reviewed by experts in the field. Additionally, these standards are being currently adopted by Canada Health Infoway and Australia’s National E-Health Transition Authority (NEHTA). Hence, it is important for NCI CBIIT to join the adoption activity at this time and provide the leadership in the area.

The guidelines on the use of the emerging ISO 21090 data type standard for projects funded by CBIIT have been approved by the Enterprise Composite Architecture Team (ECAT).

SAIF

SAIF (The HL7 SOA-Aware Interoperability Framework) provides HL7 with an Interoperability Framework, i.e. a set of constructs, best practices, processes, etc. that enable HL7 specifications to achieve cross-specification consistency and coherency irrespective of the chosen interoperability paradigm (messages, documents, or services).

SAIF consists of four core "frameworks":

1. Information (including RIM, data types, vocabulary bindings, etc.),
2. Behavior (subsuming the existing Dynamic Model),
3. Enterprise Conformance and Compliance (including HL7's existing Implementation and Conformance standards), and
4. Governance.

SAIF should be regarded as an adjunct to any EAF which focuses on Working Interoperability (WI). It is a framework which brings from SOA practice, two critical constructs which significantly enhance the path to WI, i.e.

1. Separation of concerns (static vs behavioral semantics)
2. Formal notion of contracts

SAIF is Interoperability Paradigm neutral (documents, messages, services) .

Intended use of SAIF include Enterprise interoperability projects including those building large-scale integrated health IT infrastructures at the national level.

The main benefits that will be derived from developing specifications using the SAIF include: consistency of specifications, enhanced ability to manage loosely-coupled complex interactions between multiple trading partners, increased cross-organization reuse of architecture primitives (realizing the value proposition of CMETs and extending that proposition to include behavioral as well as static semantic constructs).

Implementations include:

• National Cancer Institute (NCI),
• Canada Health Infoway (CHI),
• Open Health Tools (OHT) Architecture Project team

caBIG® Semantic Infrastructure v2 - Initiatives

The initiatives proposed below are intended to support production operation of caBIG semantics while evolution toward a national-scale capability begins. In addition to its legacy obligations, caBIG® and caGRID® are expected to inform the design and initial implementations supporting personalized medicine (BIGHealth), improving health care quality, value and affordability (the National Health Informatics Network (NHIN)) and more immediately cancer-oriented initiatives such as the cancer aware extension to the national standard electronic medical record (caEMR), and a number of other rapidly emerging initiatives that will require capabilities well beyond our traditional focus on cancer and on static data semantics.

The need for all semantic metadata to be formally recorded in a single central repository would limit or preclude application of the semantic infrastructure to very large, diverse communities such national health care. Distributed, federated metadata resources will clearly be required. Also, support for behavioral semantics describing the business context of messages and services contracts, metadata creation via processing of line of business artifacts, support for ontology based semantics and support for semantically aware service oriented architecture capabilities and some form of Grid-to-Grid semantic interoperability appear to all be capabilities that the emerging semantics infrastructure will have to provide in order to scale to national levels.

The caBIG Semantic Infrastructure and its operational model will be extended to provide integrated support for:

• Initiative 1. Distributed, federated metadata repositories and model repositories and operations.
• Initiative 2. Automated generation of metadata from line-of-business artifacts.
• Initiative 3. Rules management and contracts support (behavioral semantics)
• Initiative 4. Semantics support for W3C service oriented architecture resources.
• Initiative 5. HL7 CTS II/ OMG MIF compliant federated terminology services.
• Initiative 6. Controlled biomedical terminology, ontology and metadata content.
• Initiative 7. Assessment of semantic unification of compositional and derivational models