Experimental Platform for Model-Integrated Clinical Information Systems
MICIS - A Model-Integrated Clinical Information System
1
MICIS - A Model-Integrated Clinical Information System Janos Mathe, Jan Werner, Yonghwan Lee, Akos Ledeczi, Brad Malin, Janos Sztipanovits Institute for Software Integrated Systems, Vanderbilt University April 2-3, 2008 Enterprise Service BUS User Interface Layer Enterprise Service BUS Front End (Webpage) Application B Web Services Container Web Service Application C Front End (GUI) Application A Execution Engine Orchestration Logic Work- flows Work- Front End (Webpage) Modeling Integration Platform (MICIS-MIP) Model Transformation Layer Execution Env. Translator Policy Translator Modeling Environment Metamodel Translator Model Editor Metamodel Editor Metamodels User Models Verification Tool Translator Verification Tool Component Integration Platform (MICIS-CIP) Execution Environment CIS-DB Web Service Web Service Web Service CIS Orchestrator (CIS-O) Execution Environment Execution Engine Orchestration Logic Work- flow Work- flow Work- flow MICIS-PROPER Policy Decision Point (PDP) Policy Set (Prolog Rules) Policy Engine Policy Enforcement Point (PEP) Patient Portal Modeling Integration Platform (MICIS-MIP) Model Transformation Layer Policy Translator Modeling Environment Model Editor Policy Models Data Models Workflow Models Web Service Container (Axis2 running on Tomcat) Policy Decision Point (PDP) Policy Set Policy Store Policy Engine Policy Enforcement Point (PEP) Context Handler Web Service Web Service Policy Description Web Service Web Service Policies Policy representation, decision and enforcement • Modular Policy Enforcement Point integrated into a Web Service container • PEP supports XACML and Prolog based Policy Decision Points • Policies and policy descriptions generated from the models • Support of past events using stored facts • Support of future events using obligation mechanism Architecture of MICIS Goal Develop a Modeling and Simulation Platform for Health Information Systems Platform is suitable for • Modeling and model-based integration of Health Information Systems (HIS) - including Patient Portals (PP) - providing access to Electronic Medical Records (EMR) • Performing security and privacy analysis using model verification and simulation- based testing • Providing mapping and deployment to standard SOA execution platforms Results • Built a testbed for Health Information Systems , which includes a SOA-based execution environment, and is extended with policy evaluation and enforcement capabilities using an embedded PROLOG engine. • Developed and evaluated experimental modeling languages for three different CIS applications . • Developed model translators for translating architecture models into PROLOG clauses to enable context dependent evaluation of privacy and security policies . • Built a component library of web services for prototyping experimental HISs. In-home Patient Monitoring Sepsis Treatment Controller Deployment Architecture SOA Execution (A) Apache ODE v1.1 (or higher) Axis2 v1.3 (or higher) Tomcat Apache v5.5 (or higher) SOA Execution (B) Oracle BPEL Process Manager v10g Policy Enforcement (A.1) SWI-Prolog Axis2 handler Policy Enforcement (A.2) SunXACML Axis2 handler Policy Enforcement none Operating System Windows XP (or higher) or Linux (e.g. Fedora Core 8 / RHEL5) Hardware Regular desktop PC Approach 1) Development of abstractions in Domain- Specific Modeling Languages (DSMLs) 2) Construction of the models: capturing the key elements of operation 3) Translation (interpretation) of models 4) Execution and simulation of models Refining the level of abstraction in iterations • Integration of medical applications that use sensitive medical information • Applications communicate by using web services standards (OASIS, WC3) • Utilization of already available services is possible Applications Layer
Transcript of MICIS - A Model-Integrated Clinical Information System
MICIS - A Model-Integrated Clinical Information SystemJanos Mathe, Jan Werner, Yonghwan Lee, Akos Ledeczi, Brad Malin, Janos Sztipanovits
Institute for Software Integrated Systems, Vanderbilt University
April 2-3, 2008
Component Integration Platform (MICIS-CIP)
User Interface Layer
Exec
uti
on
En
viro
nm
ent
CIS-DB
WebService
WebService
WebService
Enterprise Service BUS
Enterprise Service BUS
Front End(Webpage)
Ap
plic
atio
n B
Web Services Container
Web Service
CIS Orchestrator (CIS-O)Execution Environment
Execution Engine
Orchestration Logic
Work-flow Work-flow Work-flow
Ap
plic
atio
n C
Front End
(GUI)
Ap
plic
atio
n A
Exec
uti
on
En
gin
e
Orc
hes
trat
ion
Lo
gic
Work-flows
Work-flows
Front End(Webpage)
Modeling Integration Platform (MICIS-MIP)
ModelTransformation
Layer
Verification ToolTranslator
VerificationTool
Exec
uti
on
En
v.Tr
ansl
ato
rP
olic
yTr
ansl
ato
r
Modeling Environment
MetamodelTranslator
Model Editor
MetamodelEditor
Metamodels
User Models
MICIS-PROPER
Policy Decision Point(PDP)
Policy Set(Prolog Rules)
Policy EnginePolicy Enforcement Point (PEP)
Enterprise Service BUS
User Interface Layer
Enterprise Service BUS
Front End(Webpage)
Ap
plic
atio
n B
Web Services Container
Web Service
Ap
plic
atio
n C
Front End
(GUI)
Ap
plic
atio
n A
Exe
cuti
on
En
gin
e
Orc
hes
trat
ion
Lo
gic
Work-flows
Work-flows
Front End(Webpage)
Modeling Integration Platform (MICIS-MIP)
ModelTransformation
Layer
Exec
uti
on
En
v.Tr
ansl
ato
rP
olic
yTr
ansl
ato
r
Modeling Environment
MetamodelTranslator
Model Editor
MetamodelEditor
Metamodels
User Models
Verification ToolTranslator
VerificationTool
Component Integration Platform (MICIS-CIP)
Exec
uti
on
En
viro
nm
ent
CIS-DB
WebService
WebService
WebService
CIS Orchestrator (CIS-O)Execution Environment
Execution Engine
Orchestration Logic
Work-flow
Work-flow
Work-flow
MICIS-PROPER
Policy Decision Point(PDP)
Policy Set(Prolog Rules)
Policy EnginePolicy Enforcement Point (PEP)
Patient Portal
Modeling Integration Platform (MICIS-MIP)
ModelTransformation
Layer
Po
licy
Tran
slat
or
Modeling Environment
Model Editor
PolicyModels
DataModels
WorkflowModels
Web Service Container(Axis2 running on Tomcat)
Policy Decision Point(PDP)
Po
licy
Set
PolicyStore
Policy Engine
Po
licy
Enfo
rce
men
t P
oin
t (
PEP
)C
on
text
Han
dle
r
WebService
WebService
PolicyDescription
WebService
WebService
PoliciesPolicy representation, decision and enforcement
• Modular Policy Enforcement Point integrated into a Web Service container• PEP supports XACML and Prolog based Policy Decision Points• Policies and policy descriptions generated from the models• Support of past events using stored facts• Support of future events using obligation mechanism
Architecture of MICIS
GoalDevelop a Modeling and SimulationPlatform for Health Information Systems
Platform is suitable for
• Modeling and model-based integration ofHealth Information Systems (HIS) -including Patient Portals (PP) - providingaccess to Electronic Medical Records (EMR)
• Performing security and privacy analysisusing model verification and simulation-based testing
• Providing mapping and deployment tostandard SOA execution platforms
Results• Built a testbed for Health Information
Systems, which includes a SOA-basedexecution environment, and is extendedwith policy evaluation and enforcementcapabilities using an embedded PROLOGengine.
• Developed and evaluated experimentalmodeling languages for three different CISapplications.
• Developed model translators for translatingarchitecture models into PROLOG clausesto enable context dependent evaluation ofprivacy and security policies.
• Built a component library of web servicesfor prototyping experimental HISs.
In-home Patient MonitoringSepsis Treatment Controller
Deployment ArchitectureSOA Execution (A)
Apache ODE v1.1 (or higher)Axis2 v1.3 (or higher)
Tomcat Apache v5.5 (or higher)
SOA Execution (B)
Oracle BPEL Process Manager v10g
Policy Enforcement(A.1)
SWI-PrologAxis2 handler
Policy Enforcement(A.2)
SunXACMLAxis2 handler
Policy Enforcement
none
Operating System
Windows XP (or higher) or Linux (e.g. Fedora Core 8 / RHEL5)
Hardware
Regular desktop PC
Approach1) Development of abstractions in Domain-
Specific Modeling Languages (DSMLs)
2) Construction of the models: capturingthe key elements of operation
3) Translation (interpretation) of models
4) Execution and simulation of models
Ref
inin
g t
he
leve
l of
abs
trac
tion
in i
tera
tion
s
• Integration of medical applications that usesensitive medical information
• Applications communicate by using webservices standards (OASIS, WC3)
• Utilization of already available services ispossible
Applications Layer
