17th International HL7 Interoperability Conference (), Standardization for Mastering Healthcare Transformation - Challenges and Solutions Prof. Dr. habil., FACMI, FACHI, FHL7, FEFMI, MIAHSI a,b a,, Germany b HL7 Germany
Organizational Organizationcentric care Process-controlled care (DMP) Person-centric care Standardization for Mastering Healthcare Transformation - Challenges and Solutions Paradigm Changes in Health Systems For increasing safety and quality of care as well as efficiency of care processes under the known demographic, social, and economical constraints, healthcare systems are - administratively since the early twenties of the last century - a matter of transformations. Administrative Managed care Health Maintenance Organizations (MHOs) (focus on process efficiency and efficacy for costs containment) with specializations disease management and case management Value-based medicine by measuring health outcome as part of care delivery, so aligning clinical practice and life sciences research get aligned Translation medicine: developing and implementing better prevention, new diagnostic tools and treatments by using a multi-disciplinary, highly collaborative process including population health, so practicing the from bench-to-bedsideto-community approach
Organizational Organizationcentric care Process-controlled care (DMP) Person-centric care Standardization for Mastering Healthcare Transformation - Challenges and Solutions Paradigm Changes in Health Systems For increasing safety and quality of care as well as efficiency of care processes under the known demographic, social, and economical constraints, healthcare systems are - administratively since the early twenties of the last century - a matter of transformations.. Methodological General care addressing health problems (one solution fits all) Phenomenological Approach Dedicated care (stratification of population for specific clinically relevant conditions) Evidence-Based Medicine Personalized, preventive, predictive and participative care considering individual health state, conditions and contexts, OMICS data, etc. (stratification of population by risk profiles) Systems Medicine, from art to multi-disciplinary science, from elementary particle to society Cognitive care
Organizational Organizationcentric care Process-controlled care (DMP) Person-centric care Standardization for Mastering Healthcare Transformation - Challenges and Solutions Paradigm Changes in Health Systems For increasing safety and quality of care as well as efficiency of care processes under the known demographic, social, and economical constraints, healthcare systems are - administratively since the early twenties of the last century - a matter of transformations.. Methodological General care addressing health problems (one solution fits all) Phenomenological Approach Dedicated care (stratification of population for specific clinically relevant conditions) Evidence-Based Medicine Personalized, preventive, predictive and participative care considering individual health state, conditions and contexts, OMICS data, etc. (stratification of population by risk profiles) Systems Medicine, from art to multi-disciplinary science, from elementary particle to society Cognitive care Technological Mainframe (KB) Client/Server (MB) Internet (GB) Distributed systems, mobile technologies, nano- and molecular technologies, knowledge representation & - management, KI, Big Data & Business Analytics, NLP, Cloud Computing, Social Business (PB, YB), Cognitive Computing
Provider Policy Provider Organization Provider Subject of Care Org. Schema Business Framework
Provider Policy Provider Organization Bridged Policy Care Process Environment, Context and Conditions Subject of Care Policy Provider Subject of Care Org. Schema Business Framework Business Process Details Individual Environment, Context and Conditions Social Environment, Context and Conditions
Care Paradigm Number of policy domains Stakeholder groups Participation of subject of care Personalized, predictive, preventive, precision medicine care dynamic Process-controlled care Organization-centered care static Paternalism Regulation Complexity Flexibility Scalability Openess
phealth & Cognitive Care Requirements
A Practical Personalized Ubiquitous Health Services Definition phealth or personalized health allows to respond to the subject of care s unique needs independent of time and location. Hence, phealth is usually defined as individually tailored health services for prevention and lifestyle, diagnosis, and treatment based on the individual assessment of personal health risks according to the individual health status, individual as well as family history, environmental and social context of the subject of care, thereby using its genetic, proteomic, anatomical, physiological and any other clinical as well as biological information. It deploys specific biomarkers, genetics, genomics, regenerative medicine and stem cell technology, liquid biopsies, etc. It exploits among others wearable and implantable sensors and actuators, bio- and molecular technologies, data integration and analytics, artificial intelligence as well as social businesses.
A Practical Personalized Ubiquitous Health Services Definition Such services require cooperation of many different and sovereign stakeholders in a multi-disciplinary approach including medicine, natural sciences, engineering, but also social and legal sciences and the entire systems sciences world (systems medicine, systems biology, systems pathology, etc.), performed through any type of principals (person, organization, device, application, component, object), deploying domainspecific methodologies, terminologies and ontologies. Such approach requires the explicit and formalized representation of involved knowledge and skills as well as the application of pervasive and autonomous computing technologies for healthcare.
phealth Objectives, Characteristics and Methodologies/Technologies Objective Characteristics Methodologies/Technologies Provision of health services everywhere anytime Openness Distribution Mobility Pervasiveness Ubiquity Wearable and implantable sensors and actuators Pervasive sensor, actuator and network connectivity Embedded intelligence Context awareness
phealth Objectives, Characteristics and Methodologies/Technologies Objectives Characteristics Methodologies/ Technologies Individualization of the system according to status, context, needs, expectations, wishes, environments, etc., of the subject of care Flexibility Scalability Cognition Autonomy Adaptability Self-organization Subject of care involvement Subject of care centration Personal and environmental data integration and analytics Service integration Context awareness Knowledge integration Process and decision intelligence Presentation layer for all actors Orchestration layer Automation, machine learning
phealth Objectives, Characteristics and Methodologies/Technologies Objectives Characteristics Methodologies/ Technologies Integration of different actors from different disciplines/domains (incl. the participation/ empowerment of the subject of care), using their own languages, methodologies, terminologies, ontologies, thereby meeting any behavioral aspects, rules and regulations Architectural framework End-user interoperability Management and harmonization of multiple domains including policy domains Reference architecture Terminology and ontology management and harmonization Knowledge harmonization Language transformation/ translation
Standardization for Mastering Healthcare Transformation - Challenges and Solutions phealth Objectives, Characteristics and Methodologies/Technologies Objectives Characteristics Methodologies/ Technologies Usability and acceptability of phealth solutions Preparedness of the individual subject of care Consumerization Subject of care empowerment Subject of care as manager Information based assessment and selection of services, service quality and safety as well as trustworthiness Lifestyle improvement and Ambient Assisted Living (AAL) services Security, privacy and trust framework Tool-based ontology management Individual terminologies Individual ontologies Tool-based enhancement of individual knowledge and skills Human Centered Design of solutions User Experience Evaluation Trust calculation services
Interoperability Challenge Communication and cooperation in dynamic, highly distributed, heterogeneous systems sets special demands on interoperability between all actors (persons, organizations, devices, applications, components, objects) involved. Interoperability describes motivation, willingness, ability, and capability to cooperate for achieving common goals or business objectives. Interoperability requires knowledge, abilities and skills, shared and adapted a- priori or dynamically at runtime, for establishing adequately cooperating associated systems.
Interoperability Challenge As mentioned before, interoperability cannot be isolated from business processes and objectives. In ubiquitous, personalized healthcare, business cases and related policies cannot be predefined any more, but are determined by state, requirements, wishes and expectations of the subject of care, thereby frequently playing the health manager role. So, interoperability services have to be adopted and provided at runtime. Differences in meeting those interoperability requirements lead to different interoperability levels. ICT systems interoperability usually addresses interoperability requirements of ICT facilities.
Healthcare as Business Domain For successfully establishing a business, you have to define: Stakeholders involved Business environment and context Business concepts Domains involved or impacting the business For planning, establishing and maintaining a business process, you have to define, to specify and to implement: Business case Business objectives Business activities and actions Stakeholder involvement Stakeholder agreements Parameters defining, impacting and controlling the business Harmonization and management of knowledge and skills
Business Concepts Definition The business concepts of a business process must be derived from the business domain s ontology. An ontology is a formal explicit specification of a shared conceptualization of a domain of interest (after Gruber). For representing a business concept, we need: Symbols / data Syntax / grammar Semantics / meaning Pragmatics / process information, purpose, evaluation criteria A business concept is specified by: Term / concept name (for implementation combined with a concept ID Concept attributes Concept meaning Relations and functions / operations, constraints
Security, Privacy and Trust Challenges of phealth Security, privacy and trust are crucial for phealth, as they influence usability and acceptability of personalized health solutions. Organizational, methodological and technological paradigm changes characterizing phealth lead to paradigm changes regarding security, privacy, and trust. Meanwhile, 50% of the ehealth spending in developed countries is dedicated to security, privacy and trust. HL7 International responds to that challenge with a series of specifications. Those standards and specifications are more mature than others because of: the close relationship to most advanced ISO specifications; the evolutionary and adaptable approach of the specifications; the deployment of an ontology-based, policy-driven approach, deploying the ISO Interoperability Reference Architecture Model.
Services Crucial for phealth Systems Care Characteristics KR and KM Characteristics Security, Privacy and Trust Characteristics Distributed care paradigm Distributed knowledge mgnt. Distributed security & privacy mgnt. Integrated care Mobile health Big data and analytics Adaptive systems Integrated framework for decision mgnt. Big data and analytics for decisions Adaptive knowledge and decision mgnt. Mobile security Big data and analytics for security & privacy Adaptive security and privacy mgnt. Person-centered care Personal decisions Personal policies Business intelligence Decision intelligence Security intelligence
Interoperability Challenge
Standardization for Mastering Healthcare 5 Transformation - Challenges and Solutions 4 3 APP DR IF 2 1 IF DR APP BC 0 ICT Ontology BC Domain Ontologies Information Perspective Domain Ontologies Organization Perspective Interoperability Level Instances Interoperability Level Technical I. Technical Plug&Play, signal & protocol compatibility Light-weight interactions Structural I. Syntactic I. Semantic I. Simple EDI, envelopes Messages and clinical documents with agreed vocabulary Advanced messaging with common information models and terminologies Information sharing Coordination Organization/Service I. Common business process Agreed cooperation Knowledge based I. Multi-domain processes Cross-domain cooperation Skills based I. Individual engagement in multiple domains Moderated end-user collaboration
Interoperability Level Required Administrative Power Required a-priori Knowledge Harmonization Standardization for Mastering Healthcare Transformation - Challenges and Solutions Simplified Interoperability Continuum
Interoperability Levels Standards Classification and Orientation Communication standard Connectivity challenge Technical orientation Communication standard Data interchange challenge Communication protocol orientation Communication standard Information exchange challenge Semantic orientation Interoperability standard Service functional cooperation challenge Operational orientation Interoperability standard Domain knowledge based interoperability challenge Intelligent systems orientation Interoperability standard Comprehensive cooperation challenge Personalized systems orientation
The Modeling Challenge of Transformative Healthcare Systems Data modeling is described as a series of processes to define data requirements for supporting business processes to meet the business objectives. Data modeling often focuses on conceptual, logical and physical data definition each representing the informational components of the organization at differing levels of abstraction [R.S. Seiner, IDERA]. The system represented by the subject of care and the processes analyzing and managing his/her health comprises all levels of granularity from atoms through molecules, cell components, cells, tissues, organs, bodies, communities, up to population. Regarding the functional, or in general inter-relational, aspects of that system, the relations comprise quantum-mechanical effects in the nano-world, biochemical processes, interrelations based on classical physics, and finally social interrelations in the macro-world. As we can consistently model and compute only systems of reasonable complexity, the system analysis or design has to address partial systems when considering higher granularity levels of the system in question.
Domain-Domain Interoperability Requesting a Permanent Bilateral Harmonization Process
Interoperability through Model and Ontology Adaptation
Interoperability Challenge under the New Organizational, Methodological and Technological Paradigms It is impossible to represent the highly complex, highly dynamic, multidisciplinary/multi-domain healthcare system by one domain s terminology or even by using ICT ontologies (such as archetypes, HL7 RIM, Zackman Framework, etc.) There are approaches for representing multi-domain concepts in an hierarchical set of ontologies. For representing advanced interoperability settings, different representations must be linked to the same real world component. For that reason, a abstract and generic reference architecture able to represent any viewpoint or domain of interest is needed. The mathematical language of Universal Type Theory and its representation by a Parameterized Barentregt Cube provides a proper solution of that challenges enabling to represent any formal or informal language. Current approaches claiming to solve that problem do this on the basis of implicit knowledge or by using ICT representation tools the addressed domain experts, which should be in the lead, cannot understand. The abstract representation of the universe provide an alternative.
Universal Type Theory Basis of the GCM Barendregt Cube with Parameters After Kamareddine et al.
Interoperability Mediated by the Interoperability Reference Architecture Model
GCM Granularity Levels
The Implementation Challenge After developing an abstract system s architecture, it must be instantiated for concrete business domains. The resulting real system components must be properly named and described, using pre-existent terminologies and ontologies where possible. To get closer to the vision of comprehensive interoperability, the ontological representations used by different domain experts for representing entities in reality must be harmonized. For that purpose, the ontological representation must be provided at a level of formalization and expressivity which guarantees common understanding, i.e. expresses meaning and rules as explicit as needed depending on education, skills, and experiences of the actors involved.
Business View Enterprise View Information View Computational View Engineering View Technology View System s Architectural Perspective System Component Composition Standardization for Mastering Healthcare Transformation - Challenges and Solutions Domain n Domain 2 Domain 1 Domain Perspective Business Concepts Relations Networks Aggregations Details Development Process Perspective System Viewpoint
A practical implementation of the universal description of complex systems covering any domain is a system-oriented, architecture-centric, ontology-driven Generic Reference Model (GCM), which finally allows the automated mapping of any model. The integration of the systems and the GCM ontologies into BFO and OBO is ongoing. The feasibility of the approach has been demonstrated for mapping/transformation of HL7 v2 and HL7 v3. Another example is the automated development of interoperable Webservices for Type 2 Diabetes Care Settings including primary, secondary, and tertiary care, home care and self-engagement, dieticians, etc., based on the proposed approach. Furthermore, this approach is basis for a series of ISO and HL7 standards, as presented in the following slides.
System s Architectural Perspective Business VP Enterprise VP Information VP Computational VP Engineering VP Technology VP System Component Composition System s Architectural Perspective Standardization for Mastering Healthcare Transformation - Challenges and Solutions Domain an Domain a2 Domain a1 Domain Perspective Business Concepts Relations Networks Aggregations Development Process Perspective Details Domain n Domain 2 Domain 1 Business Concepts Relations Networks Aggregations Details System Viewpoint Domain bm Domain b2 Domain b1 Domain Perspective Business Concepts Relations Networks Aggregations Details Development Process Perspective
Service Task Healthc. Serv. Medical Unit Medical Care Funct. Role Encoding Decoding Deleg. Decis. Auth. Decis. Access Policy Act. Subst. ID Mgmt. Authenticat. E-Sign. Serv. Audit Service Workplace Component Refrain Policy Oblig. Policy Deleg. Policy Auth. Policy Person Application Device Drug Equipment Appl. Security Comm. Sec. Room Safety Basic Policy Comp. Policy Struct. Role Organization Mean Inf.Sec. Mgmt. Build., Area Entry. Composition Folder Rec. Comp. Mess. Item Mess. Comp. EHR-Message Comm. Type Policy Actor Facility Location Resource: Location Resource: Facility Resource: Actor Policy Security Services Medical Care Business Concepts Standardization for Mastering Healthcare Transformation - Challenges and Solutions Re-engineering of the ISO 13606 Reference Model Related Party Relations Network Aggregations Details Business Viewpoint Inf. VP Eng. VP
Conclusions 1 Interoperability is not first a matter of the ICT domain, but a user domains one. Interoperability is a challenge to meeting business objectives. It requires sharing of knowledge and skills, which should be built on a hierarchical system of ontologies. Interoperability must go beyond simplification and abstraction for data integration at ICT ontology level towards concept and rule harmonization and integration at business domains ontologies level. Multi-disciplinary interoperability solutions (life sciences, natural sciences, technology, legal and social sciences, etc.) require an architecture-centric systems approach to the domains of discourse represented by their ontologies, so enabling the formalization of systems representation and integration including ontology mapping. Flexible, scalable, business-controlled, adaptive, knowledge-based, intelligent systems must follow a systems-oriented, architecture-centric, ontology-based and policy-driven approach.
Conclusions 2 The modeling process is an iterative one where the first iteration is performed in a top-down manner to guarantee the conceptual integrity of the model. The view on the model as well as the way of structuring and naming the concepts of the problem space have to be defined by the relevant stakeholders. The business domain drives the modeling of interoperability solutions. So, domain experts are crucial for interoperability standards development. Interoperability standardization efforts have to go beyond the traditional health information system perspective towards a comprehensive reflection of the business domain. Closer cooperation amongst the different SDOs (also beyond the field of health informatics), but also with domain specific professional associations and societies is inevitable.
Thank you very much for your kind attention! Contact: Prof. Dr. habil., FACMI, FACHI, FHL7, FEFMI, MIAHSI, Visiting Professor, First, Charles University Prague, Czech Republic c/o HL7 Germany An der Schanz 1 50735 Köln Germany Email: bernd.blobel@klinik.uni-regensburg.de