Download presentation
Presentation is loading. Please wait.
1
0 CBIIT’s Technology Strategy: Moving to a Semantically-Aware Services-Oriented Architecture (sSOA) Charlie Mead, MD, MSc Chief Technology Officer Center for Biomedical Informatics & Information Technology
2
1 21st Century Biomedical Paradigm: A “Learning Health System” Personalized, Predictive, Pre-emptive, Participatory… Unifies discovery, clinical research, and clinical care (bench-bedside-back) into a seamless continuum Accelerates the time from discovery to patient benefit Enables a health care system, not disparate “sectors” or “silos” Empowers consumers to manage their health over a lifetime Results in improved clinical outcomes
3
2 The “integrated/linked” NCI-supported Cancer Community: a complex system with diverse information exchange and behavior coordination requirements for interoperability
4
3 Cancer Knowledge Cloud Consumer Community Hospital Research Hospital Research Institution Comparative Effectiveness Quality Pharmaco- Vigilance Industry Decision Support Data Aggregators Analytics Biomedical Research Bio- Surveillance Practice The “Cancer Knowledge Cloud”
5
4 Focus on Interoperability: Describing Complex Systems From NCICB caBIG™ to CBIIT BIG Health: –More stakeholders –More agendas –More complexity Multiple vertical levels of organization and horizontal processes which – in order to produce the system’s products-of-value for its stakeholders/user – cross vertical (interface) boundaries. (adapted from Ivar Jacobson) Experience in multiple domains has repeatedly shown that complexity is most effectively managed through layering and use of standards at the “interface boundaries.”
6
5 Consume/ContributeValidateInform CBIIT Development: Bringing Standards to the Community 5 CBIIT consumes and develops conformant standards-based specifications to resolve business problems CBIIT validates the applicability of specifications via reference implementations CBIIT informs the commercial and open-source vendor communities of the “state of the art” by deploying/handing off its reference implementations.
7
6 NCI sSOA: Utilizing Standards NCI Services are based, wherever possible, on existing standards to enhance interoperability –Payloads are derived from the HL7 v3 Reference Information Model (RIM) wherever possible BRIDG now formally mapped to RIM –Attributes bind to ISO 21090 data types –Utilize standard controlled biomedical terminologies, e.g. LOINC Mappings maintained via NCI Thesaurus and the ISO 11179 metadata specification
8
7 1.Enterprise-Level Interoperability –Our applications, data, and analytic services need the capacity to connect and meaningfully exchange information and coordinate behavior. CBIIT “BIG 3” Priorities for 2010 - 2011
9
8 2.Enterprise Security –Develop and implement a clear, integrated enterprise strategy and operational tactics that unify and integrate our approach to security. CBIIT “BIG 3” Priorities for 2010 - 2011
10
9 3.Deployment Support –Comprehensively support the requirements of the 21 st -century healthcare community, including On-going stakeholder identification and management; and Development of technology-independent specifications and reference implementations of software that provides clear benefit to that community. CBIIT “BIG 3” Priorities for 2010 - 2011
11
10 Preview/Press Release Evolution, not Revolution –Architecture must support business Problem Statement: CBIIT needs to… –define, design, develop, and deploy software components defined as business capabilities rather than “monolithic applications;” –certify software components at a more granular level than the current “metal level” approach; –enable automated/semi-automated certification; and –provide technology-independent specifications that enable organizations to build software components that will interoperate with CBIIT components. 10
12
11 Semantically-Aware Services-Oriented Architecture (sSOA) Semantics are critical to interoperability in the domain of healthcare, clinical research, and life sciences –“bedside-to-bench-and-back-and-beyond” == computable semantic interoperability What is a SOA? –“An technology architecture Design Paradigm which can be applied in the context of Distributed Computing which collectively expresses a set of core Design Principles which collectively realize a set of Design Characteristics?” -- Thomas Erl, “Principles of Service Design” caBIG 1.x is not an sSOA –Not a value judgment, just a statement of fact –“That was then, this is now…” –“Our baby is not ugly…it’s just growing up.” 11
13
12 SOA Strategic Goals (CBIIT “semantics” frame these Goals) Intrinsic Interoperability –Interoperability vs Integration Increased Federation –Common endpoint and local governance Increased business/technology alignment –Linear “degree of difficulty” for change Increased vendor neutrality options –Specifications at a logical level (ECCF PIM) 12 -- Thomas Erl, “Principles of Service Design”
14
13 SOA Strategic Benefits (CBIIT “semantics” frame these Benefits) Increased ROI –Reuse vs inconsistent Redundancy –Applications as “compositional aggregations” Increased Organizational Agility –Business evolution aligned with technology flexibility Reduced IT footprint –Elimination of redundancy and inconsistency –Elimination of one-off integration activities –Decreased governance burden 13 -- Thomas Erl, “Principles of Service Design”
15
14 SOA Design Principles (CBIIT “semantics” frame these Principles) Standardized Service Contracts Service Loose Coupling Service Abstraction Service Reusability Service Autonomy Service Statelessness Service Discoverability Service Composability 14 -- Thomas Erl, “Principles of Service Design”
16
15 Design Principles vs Strategic Goals and Benefits – Thomas Erl, “Principles of Service Design” 15 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X Intrinsic Inter- operability Increased Federation Increased Business/T echnology Alignment Increased Vendor Diversificati on Options Increased ROI Increased Organiza- tionalbAgility Decreased IT Burden Standard Service Contracts Service Loose Coupling Service Abstraction Service Reusability Service Autonomy Service Statelessness Service Discoverability Service Composability
17
16 SOA Challenges SDP increased design complexity SDP need for Design Standards (informational, behavioral) SDP top-down delivery requirements SDP “counter-agile” (contract-first) design/delivery SDP new organizational roles, relationships, responsibilities SDP governance requirements 16 -- Thomas Erl, “Principles of Service Design”
18
17 Preview/Press Release Advantages gained from sSOA: CBIIT users will realize benefits from the solution to each aspect of the Problem Statement:… –focus on business capabilities increased agility –fine-grained certification increased predictability and reliability –automated certification increased responsiveness to change and evolution of software components –technology-independent specifications increased vendor participation and cross-vendor interoperability 17
19
18 sSOA ServiceTaxonomy (1) NCI uses a version of the CBDI taxonomy –Placement of a given service is purely for classification purposes Infrastructure/Utility –Perform “background” essential functions required by most other services Security Transaction management/Auditing Terminology Management Core –Provide information components to Capability and Process services Person, Organization, Relationship management
20
19 sSOA Service Taxonomy (2) Capability –Provide “business atoms” from which more complex business processes/workflows are constructed Image Management Lab Management Process –Arbitrarily complex services that utilize the other three service types to carry out business functions Adverse Event Management Outcomes Management
21
20 sSOA ServiceTaxonomy (3) Each service exposes multiple –Functional Profiles (collections of operations) bound to –Semantic Profiles (specifics of data exchanged) –Together, the F and S profiles are linked to traceable business processes or requirements All specifications are documented via a series of artifacts collected in a SAIF construct called the Enterprise Conformance and Compliance Framework (ECCF) –Actual “collector” is called a “Specification Stack” –A component of the HL7 Services-Aware Interoperability Framework (SAIF)
22
21 CBIIT Road Map to sSOA Adoption/adaption of HL7 Services-Aware Interoperability Framework (SAIF) –CBIIT SAIF Implementation Guide (SAIF IG) Standardization of enterprise-level components will be required –However, standardization can be viewed as “layered” depending on its “breadth” of required interoperability ECCF “Stairway to Heaven” quantifies degree-of-difficulty required to achieve interoperability Industry standards and CBIIT Design Standards Governance Known standards that will form the basis for CBIIT-focused interoperability tapestry (e.g. caBIG®, caEHR, BIG Health) –Existing caBIG® concepts and terminologies –21090 data types 21
23
22 The SAIF/ECCF Value Proposition: (layered) Working Interoperability Working Interoperability “The deterministic exchange of data/information in a manner that preserves shared meaning (static & behavioral semantics)” A − F: Trading Partners Two trading partners interoperate based on a certified level of shared compliance to interoperability specifications/standards. Certified level-of-conformance determines the possible degree of automated interoperability and/or the difficulty of the transformations that are required to enable interoperability.
24
23 CBIIT Road Map to sSOA via SAIF/ECCF Some things will stay the same: –CBIIT focus: Support for the ongoing goals supporting information and functional integration across and throughout the Translational Medicine Continuum… “from bedside to bench and back…and beyond” –Oversight of CBIIT-funded software development 23
25
24 CBIIT Road Map to sSOA via SAIF/ECCF Some things will evolve: –Static data, metadata, terminology representations, and tooling –Application development from monolithic to service composition –Increased emphasis on enterprise architecture –Increased emphasis on Continuous Integration (aka BDA) 24
26
25 CBIIT Road Map to sSOA via SAIF/ECCF Some things will be new: –Behavioral metadata that is defined and captured –Artifacts that define ECCF Specification Stack instances –Increased emphasis on automated testing/certification –Tooling to support new data/metadata and processes –Focus on software contracts as basis for Computable Semantic Interoperability (CSI) –Formal requirements traceability –Formal project and architecture governance –Some aspects of software engineering process –Some aspects of contracts for development organizations 25
27
26 The ECCF Specification Stack
28
27 Conformance Certification Conformance Statements for a given SS are shown in horizontal black arrows. Conformance Assertions for a given Technology Binding in vertical red arrows. Conformance Statements == Testable Requirements Conformance Assertions == Implementer claims Testable via a Test Harness defined by Conformance Statements Abstraction Layers CIM Technology Binding Specification Stack Conformance Statements PIMPSM
29
28 The CBIIT ECCF Implementation Guide (IG) What it is: CBIIT’s implementation details of the HL7 “SAIF Book” (referred to as “Implementation Guide(s)” (IGs)) –Information Framework –Behavior Framework –Governance Framework (reuse of standards) –Enterprise Conformance and Compliance Framework When –ECCF IG is first Draft available ~May 15 Full operationalization by Fall 2010 –Semantic Infrastructure (v2) will include IF and BG IGs –Enterprise architecture governance SAIF GF IG 28
30
29 Compatibility: the suggestion that one behavior may align with another Interoperability: the predictable/reliable exchange or coordination of meaning Guideline: a informal indication of how things should be done which can be contextually modified Specification: an unambiguous definition of how things must be done which can be modified based on explicit rules Compatibility vs Interoperability Guideline vs Specification
31
30 Scope & Highlights Platform Specific artifacts Annotated Models (CDE’s and EVS concepts) Service Interfaces caBIG Silver Level Compatibility Reviews 12 required artifacts 4 optional artifacts caBIG Compatibility Guidelines: Today
32
31 Silver Compatibility Required Artifacts 1.PPT Presentation for Review Team (General description) 2.UML Model 3.Graphic representation of UML model 4.Semantically Annotated XMI file 5.Error Log from SIW (Semantic Integration Workbench) 6.UML Model submission form (Proof loaded into caDSR) 7.Vocabulary Report (Proof used Standard Vocabs) 8.Standards Report (Proof re-used CDE Standards for interoperability)
33
32 Silver Compatibility (2) Required Artifacts (continued) 9.Full CDE use report (Proof of CDEs in caDSR) 10.API description Doc (eg, Java Docs) 11.Test Scripts of API 12.Test logs of Interface/API functions Optional Artifacts 1.Use case Document 2.Requirements/specification document 3.Implementation/Design Document 4.Test Approach Document
34
33 caBIG Compatibility Guidelines: Today Compatibility Guidelines Today Platform Specific Annotated Models (CDE’s) Service Interfaces
35
34 ECCF “bullets” Layered Specification Testable Conformance (Boolean) Computable Behavioral and Static Semantics Requirements Traceability to explicit service specifications Binding to standards when applicable Future CBIIT Conformance Guidelines: Submission Package Non-computable artifacts (links to documents, etc.) Computable artifacts Service Metadata (BVP, IVP, CVP, EVP) Conformance Statements (testable requirements) CBIIT Interoperability Specifications: Tomorrow
36
35 CBIIT Interoperability Specifications: Tomorrow Conformance Guidelines Tomorrow Layered Specifications Testable Conformance Behavioral Semantics Traceability Binding to standard models and data types
37
36 ECCF Guidelines & Artifacts: Tomorrow Analyst/ Architect BPEL CONSOLE ECCF Specification ECCF Registry (Service Specifications) Models & Bindings & Rules CIM PIM PSM Wiki (Non Computable) Automated Conformance Testing SI Tooling LoB Artifacts Profiles/Models
38
37 Summary: The Value of an ECCF The enemy of Working Interoperability is implicit assumptions. –Implicit assumptions are always explicitly realized in an implementation. When expressed explicitly prior to the actual code, these assumptions are referred to as conformance statements. –Implementations can then make pair-wise conformance assertions against conformance statements –Conformance is thus testable, verifiable, and certifiable. The ECCF provides a framework for collecting the set of artifacts with explicitly define the complexity inherent in software components.
39
38 The BRIDG Model, V3.0 38
40
39 Applying Standards for Collaboration: Clinical Research Filtered Query Service 39
41
40 Evolution – not Revolution Data Elements to Business- Level Information I Business-Level Information provides Mature Business Capabilities II Mature Business Capabilities organize the Knowledge Cloud III Data services Business Capability Services Data models Semantic Profiles bound to service Functional Profiles
42
41 NCI Enterprise sSOA Periodic Table of Services 41 R Registration Ae Adverse Event Pt Protocol “P ROCESS ” “ C ORE ” “C APABILITY ” S Specimen Sc Scheduling Tx Translation Au Audit Ra Referral and Authorization Hx Hx and Physical Va Validation Ev Enterprise Vocabulary O Organization P Person A Agent D Disease C Correlation Pa Protocol Abstraction Dx Discharge Note Po Patient Outcomes Tp Treatment Plan Ds Decision Support I Image L Lab Rx Pharmacy “Infra / U TILITY ” Cm Contract Management Mp Master Problem List Ay Allergy Sd SDTM E Eligibility Km Knowledge Management Cr Credentialing Oc Study Outcomes Qr Data Query Id Management Tr Trust Management Aa Authorization Authentication Py Policy
43
42 NCI sSOA Periodic Table of Services: Infrastructure 42 R Registration Ae Adverse Event Pt Protocol “P ROCESS ” “ C ORE ” “C APABILITY ” S Specimen Sc Scheduling Tx Translation Au Audit Ra Referral and Authorization Hx Hx and Physical Va Validation Ev Enterprise Vocabulary O Organization P Person A Agent D Disease C Correlation Pa Protocol Abstraction Dx Discharge Note Po Patient Outcomes Tp Treatment Plan Ds Decision Support I Image L Lab Rx Pharmacy “Infra / U TILITY ” Cm Contract Management Mp Master Problem List Ay Allergy Sd SDTM E Eligibility Km Knowledge Management Cr Credentialing Oc Study Outcomes Qr Data Query Id Management Tr Trust Management Aa Authorization Authentication Py Policy Trust Management establishes verifiable and explicit relationships between technology endpoints in the deployed architecture Identity Management deals with identities of specific entities of interest to the organization Authorization and Authentication Management provides a means of testing an identity against a set of credentials and explicit policy within a deployed architecture Policy Management exposes elaborated, well-formed policies and rules that manifest the organization's permissions, prohibitions, and obligations
44
43 NCI sSOA Periodic Table of Services: Infrastructure 43 R Registration Ae Adverse Event Pt Protocol “ PROCESS ” “ C ORE ” “C APABILITY ” S Specimen Sc Scheduling Tx Translation Au Audit Ra Referral and Authorization Hx Hx and Physical Va Validation Ev Enterprise Vocabulary O Organization P Person A Agent D Disease C Correlation Pa Protocol Abstraction Dx Discharge Note Po Patient Outcomes Tp Treatment Plan Ds Decision Support I Image L Lab Rx Pharmacy “Infra / U TILITY ” Cm Contract Management Mp Master Problem List Ay Allergy Sd SDTM E Eligibility Km Knowledge Management Cr Credentialing Oc Study Outcomes Qr Data Query Id Management Tr Trust Management Aa Authorization Authentication Py Policy Service Contract Management represents a series of capabilities around the storage, versioning, and expression of the contract semantics supporting interoperability Knowledge Management represents a series of capabilities around the storage, versioning, and expression of the semantics supporting key capabilities Enterprise Vocabulary Management supports the management, storage, and mapping of terminologies and value sets Validation Management verifies structural and semantic consistency across messages used in interoperability scenarios Transformation Management provides a functional end point to manage and enact mappings between syntactically disparate information types Auditing Management provides an interface that captures auditing information around the access to sensitive data
45
44 NCI sSOA Periodic Table of Services: Core 44 R Registration Ae Adverse Event Pt Protocol “ PROCESS ” “ C ORE ” “C APABILITY ” S Specimen Sc Scheduling Tx Translation Au Audit Ra Referral and Authorization Hx Hx and Physical Va Validation Ev Enterprise Vocabulary O Organization P Person A Agent D Disease C Correlation Pa Protocol Abstraction Dx Discharge Note Po Patient Outcomes Tp Treatment Plan Ds Decision Support I Image L Lab Rx Pharmacy “Infra / U TILITY ” Cm Contract Management Mp Master Problem List Ay Allergy Sd SDTM E Eligibility Km Knowledge Management Cr Credentialing Oc Study Outcomes Qr Data Query Id Management Tr Trust Management Aa Authorization Authentication Py Policy Organization Management establishes a central service of record for the Organization information type Correlation Management establishes a central service of record for relationships between core information components (Persons to Organizations, etc.) Person Management establishes a central service of record for the Person Information Protocol Abstraction Management establishes a central service of record for non-annotated, "flat" information about Protocols and their relation to trials Disease Management establishes a central service of record for information about various diseases found in trials Agent Management establishes a central service of record for information about various agents found in trials
46
45 NCI sSOA Periodic Table of Services: Capability 45 R Registration Ae Adverse Event Pt Protocol “ PROCESS ” “ C ORE ” “C APABILITY ” S Specimen Sc Scheduling Tx Translation Au Audit Ra Referral and Authorization Hx Hx and Physical Va Validation Ev Enterprise Vocabulary O Organization P Person A Agent D Disease C Correlation Pa Protocol Abstraction Dx Discharge Note Po Patient Outcomes Tp Treatment Plan Ds Decision Support I Image L Lab Rx Pharmacy “Infra / U TILITY ” Cm Contract Management Mp Master Problem List Ay Allergy Sd SDTM E Eligibility Km Knowledge Management Cr Credentialing Oc Study Outcomes Qr Data Query Id Management Tr Trust Management Aa Authorization Authentication Py Policy Specimen Management provides the core specification for specimen management from both the clinical and research perspectives regardless of the nature of specimen transactions that occur or the type of specimen Credentialing Management allows the credentials of a given investigator to be enumerated and validated by a trusted third party Treatment Plan Management provides the ability to create and update Treatment Plan templates, generate and view prospective calendars of patient activities, track activities as they occur, and manage patient calendars as they change during a plan Image Management includes the ability to manage image, including intersections with order reporting. Lab Management includes the ability to order, track, and manage laboratory orders from a clinical perspective, including intersections with specimen management and pathology reporting Pharmacy Management includes the ability to order, track, and manage prescriptions from a clinical perspective Scheduling Management provides the capabilities associated with scheduling a particular appointment for a particular encounter
47
46 NCI sSOA Periodic Table of Services: Capability 46 R Registration Ae Adverse Event Pt Protocol “ PROCESS ” “ C ORE ” “C APABILITY ” S Specimen Sc Scheduling Tx Translation Au Audit Ra Referral and Authorization Hx Hx and Physical Va Validation Ev Enterprise Vocabulary O Organization P Person A Agent D Disease C Correlation Pa Protocol Abstraction Dx Discharge Note Po Patient Outcomes Tp Treatment Plan Ds Decision Support I Image L Lab Rx Pharmacy “Infra / U TILITY ” Cm Contract Management Mp Master Problem List Ay Allergy Sd SDTM E Eligibility Km Knowledge Management Cr Credentialing Oc Study Outcomes Qr Data Query Id Management Tr Trust Management Aa Authorization Authentication Py Policy Data Query Management provides a consistent mechanism to query and retrieve information from information systems SDTM Management provides a data structure that is reused in regulatory reporting Master Problem Management establishes a central service of record for the Problems tied to patients Allergy Management establishes a central service of record for the Allergies tied to patients
48
47 NCI sSOA Periodic Table of Services: Process 47 R Registration Ae Adverse Event Pt Protocol “ PROCESS ” “ C ORE ” “C APABILITY ” S Specimen Sc Scheduling Tx Translation Au Audit Ra Referral and Authorization Hx Hx and Physical Va Validation Ev Enterprise Vocabulary O Organization P Person A Agent D Disease C Correlation Pa Protocol Abstraction Dx Discharge Note Po Patient Outcomes Tp Treatment Plan Ds Decision Support I Image L Lab Rx Pharmacy “Infra / U TILITY ” Cm Contract Management Mp Master Problem List Ay Allergy Sd SDTM E Eligibility Km Knowledge Management Cr Credentialing Oc Study Outcomes Qr Data Query Id Management Tr Trust Management Aa Authorization Authentication Py Policy History and Physical Management provides access to patient's History and Physical document. This service will facilitate requests for and provision documents that capture the clinical history of the patient during a specific encounter Adverse Event Management provides queries, reports adverse events and the associated work flows Discharge Note Management provides access to a patient's Discharge Notes for a given encounter Decision Support provides an algorithmically-neutral way to manage decision support services for patients Referral is Management provides core capabilities needed by health care organizations to facilitate the business process of patient referrals and consultations among providers
49
48 NCI sSOA Periodic Table of Services: Process 48 R Registration Ae Adverse Event Pt Protocol “ PROCESS ” “ C ORE ” “C APABILITY ” S Specimen Sc Scheduling Tx Translation Au Audit Ra Referral and Authorization Hx Hx and Physical Va Validation Ev Enterprise Vocabulary O Organization P Person A Agent D Disease C Correlation Pa Protocol Abstraction Dx Discharge Note Po Patient Outcomes Tp Treatment Plan Ds Decision Support I Image L Lab Rx Pharmacy “Infra / U TILITY ” Cm Contract Management Mp Master Problem List Ay Allergy Sd SDTM E Eligibility Km Knowledge Management Cr Credentialing Oc Study Outcomes Qr Data Query Id Management Tr Trust Management Aa Authorization Authentication Py Policy Protocol Management supports creation of a Study Protocol by providing a capability to create a computable part of the protocol Trial Registration Managment provides investigators with the ability to submit descriptive information about a clinical trial. It also provides the ability to search registered trials based on various criteria Study Outcomes Management establishes a set of reporting patterns for collecting data and validating it in order to report on a particular intervention outcome Patient Outcomes Management provides clinicians or administrators at clinicians' offices with the ability to submit outcome data for the cancer patients. It also provides them with the ability to query the outcomes data they submitted Eligibility Management provides the capabilities associated with establishing eligibility for a patient to register on a Study based on criteria specified in the Study Protocol
50
49 NCI Enterprise sSOA: Periodic Table of Services 49 R Registration Ae Adverse Event Pt Protocol “ PROCESS ” “ C ORE ” “C APABILITY ” S Specimen Sc Scheduling Tx Translation Au Audit Ra Referral and Authorization Hx Hx and Physical Va Validation Ev Enterprise Vocabulary O Organization P Person A Agent D Disease C Correlation Pa Protocol Abstraction Dx Discharge Note Po Patient Outcomes Tp Treatment Plan Ds Decision Support I Image L Lab Rx Pharmacy “Infra / U TILITY ” Cm Contract Management Mp Master Problem List Ay Allergy Sd SDTM E Eligibility Km Knowledge Management Cr Credentialing Oc Study Outcomes Qr Data Query Id Management Tr Trust Management Aa Authorization Authentication Py Policy NCI is continually refining our periodic table of services to reflect the needs of the cancer research and care community
51
50 3/31/201050 Current State of All Services
52
51 CBIIT Service Inventory
53
52 CBIIT Service Inventory
54
53 CBIIT Service Inventory
55
54 CBIIT Service Inventory
56
55 3/31/201055 CBIIT Service Inventory
57
56
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.