+ All Categories
Home > Documents > Semantic Data Capture Initiative

Semantic Data Capture Initiative

Date post: 27-Jan-2016
Category:
Upload: gustav
View: 42 times
Download: 0 times
Share this document with a friend
Description:
Semantic Data Capture Initiative. Hemant Shah M.D., M.Surg. Sr. Research Informatician Henry Ford Health System [email protected]. What is Semantic Data Capture Initiative (SDCI)?. - PowerPoint PPT Presentation
Popular Tags:
34
Semantic Data Capture Initiative Hemant Shah M.D., M.Surg. Sr. Research Informatician Henry Ford Health System [email protected]
Transcript
Page 1: Semantic Data Capture Initiative

Semantic Data Capture Initiative

Hemant Shah M.D., M.Surg.

Sr. Research InformaticianHenry Ford Health System

[email protected]

Page 2: Semantic Data Capture Initiative

What is Semantic Data Capture Initiative (SDCI)?

A project to develop decision support capabilities combined with structured data capture for CarePlus NG, and its evaluation

Under a TATRC (DoD) supervised grant All code is available as Open Source (EPL) Period of Performance

From April 2008 to October 2010

Page 3: Semantic Data Capture Initiative

The Team Investigators:

R. David Allard, MD (PI) Hemant Shah, MD (Co-PI) Patricia Williams Ganesh Krishnan (Lead Developer)

Support from: CPNG development team Rich Vollmerhausen CSRI Research Staff CSRI Training Staff Gloria Whitten Marie McLenaghan

External Adviser: Prakash Nadkarni MD

Page 4: Semantic Data Capture Initiative

Goal

Provide Data Capture in CarePlus EMR, which is:StructuredStandard BasedSemantic

Encourage point-of-care data entry with Decision Support as an incentive for the clinician

Page 5: Semantic Data Capture Initiative

Hypothesis

Clinical documentation templates that leverage metadata with controlled medical

vocabularies and interoperate with clinical decision support can be integrated into key

ambulatory care processes in a manner acceptable to clinicians and support staff

Page 6: Semantic Data Capture Initiative

Components of SDCI

Research Component Software Development Component Knowledge Development ComponentEvaluation Component

Page 7: Semantic Data Capture Initiative

Research Component

Structured Data Entry Literature Review Guideline Systems Comparative Analysis Controlled Medical Vocabulary Analysis

Criteria specification for Vocabularies for CPNG and Decision Support

Examine Vocabularies SNOMED-CT LOINC ICD-9 CPT NCI Thesaurus UMLS Metathesaurus MEDCIN 3.0

Application of Criteria Industry Scan of existing systems

Page 8: Semantic Data Capture Initiative

Development ComponentProtean – Guideline Authoring Tool

Proteus Guideline Engine

Greed – Rule Authoring tool

Semantic Annotation Tool

Interactions with CPNGMetadata RepositoryPatient Record

Page 9: Semantic Data Capture Initiative

Editable Clinical Process

Page 10: Semantic Data Capture Initiative

Editable Clinical Process

Page 11: Semantic Data Capture Initiative

Internet

Knowledge Component Repository

Editable Clinical Process

CardiovascularSystemExpert

Diabetes

Expert

Page 12: Semantic Data Capture Initiative

Proteus may be used for…

Creating executable clinical guidelines to provide decision-support to clinicians about individual patients

Creating process-oriented EMRs with integrated clinical decision-support

Page 13: Semantic Data Capture Initiative

Proteus Model Contains…

A specification of an architecture for:KCsExecutable Workflows (Guidelines) built with KCsTools and Systems to handle them

A graphical language for Workflows Human & machine readable Proteus Graphical Language - PGL

Page 14: Semantic Data Capture Initiative

Knowledge Component (KC)

A modular building block for Clinical Processes

Each KC Represents either a Clinical Action, a Clinical Event or a Clinical Process

Contains knowledge about a clinical activity: Actions to be performed Events to look for Data to be collected from the actions and events Interpretation and implications of that data Supplementary information about the activities (e.g.

links to websites)

Page 15: Semantic Data Capture Initiative

Knowledge Component Basics

KC Represents: Clinical Process (e.g. diagnosis

of acute abdomen pain)

Clinical Atomic Activity, which may be: Clinical Action (e.g. palpation of

liver)

Clinical Event (e.g. vomiting)

Knowledge Component (KC)

KC may contain data-fields describing the underlying clinical entity

Lump

Tenderness

Vomiting

Temperature

AbstractionAbstraction

Value of KC

Page 16: Semantic Data Capture Initiative

Knowledge Component Features

KCs can be NestedTo represent composite processesTo reduce complexity

KCs can be linked by Activity-links To represent processes To define guidelines

LumpTendernessVomitingTemperature

Instantiated (executed) KCs become medical record

Abstraction

yessevereyes

102 F

Page 17: Semantic Data Capture Initiative

Inference Tools as Components (SOA)

Only a reference to an Interface

Swappable

Location neutral

Inferencing technology neutral

Technologies•Hard Coded Algorithm•Production Rules •Decision Tables•Decision Theory•Neural Networks •Fuzzy Logic •Any other …

•Patient assisted decisions•Human expert (even user)•Combination of these

Technologies•Hard Coded Algorithm•Production Rules •Decision Tables•Decision Theory•Neural Networks •Fuzzy Logic •Any other …

•Patient assisted decisions•Human expert (even user)•Combination of these

Page 18: Semantic Data Capture Initiative

Inference Tools as Components

Two Types of Inference Tools

Inference tool for Abstraction

Inference tool for Action

Page 19: Semantic Data Capture Initiative

Inference Tool for Abstraction

Decides Abstraction – The

value of the KC

Lump

Tenderness

Vomiting

Temperature

Abstraction

Page 20: Semantic Data Capture Initiative

Inference Tool for Action

Decides If an Action has to be

triggered based upon some intelligence

Test ATest A

Test BTest B

Test CTest C

Action AAction A

Action BAction B

Page 21: Semantic Data Capture Initiative

Protean – An Integrated Process Authoring Tool

A Demo

Page 22: Semantic Data Capture Initiative

Data

ElementsPatient Data

Vocab Vocab ServerServer

Vocab Vocab ServerServer

Patient Data

Patient DataSemantic Data Elements

Semantic Data Elements

TemplateTemplate

ProteanProteanProteanProtean

CarePlus NGCarePlus NG

ViewViewViewViewEHR Adaptor EHR Adaptor (vMR)(vMR)

EHR Adaptor EHR Adaptor (vMR)(vMR)

Proteus Proteus Guideline Guideline

EngineEngine

Proteus Proteus Guideline Guideline

EngineEngine

GreEd GreEd GreEd GreEd

Rule Rule EngineEngine

Rule Rule EngineEngine

Guideline Guideline RepositoryRepository

Guideline Guideline RepositoryRepository

Semantic Semantic Annotation Annotation

ToolTool

Semantic Semantic Annotation Annotation

ToolTool

Semantic Data Capture Initiative – Architecture

SNOMED-CT

Page 23: Semantic Data Capture Initiative

What is Greed? A tool to author and edit rules

Easy to use graphical representation of rules Drag and drop is all you need Internal rule syntax inspired by LISP Ability to create rules in multiple languages, e.g.,

Arden Syntax, Java, RuleML, Jess, JBoss Rules etc. Semantic and Completeness checks on rules Allows testing of rules from within the environment Currently in use by Protean Future Plans:

Use ISO/IEC 11179 data elements for conditions and inferences

Extensibility – New logical or math operations can be added Rule repository related features To be made available as an independent Rules system

Page 24: Semantic Data Capture Initiative

Semantic Annotation Tool Allows authoring of standard based data elements that

are linked (annotated) with concepts in SNOMED-CT Connects to CPNG metadata repository and gets ‘raw’

data elements Allows authors to select appropriate concepts for them

and annotate the data elements with them Stores the annotated data elements which are now

called Semantic Data Elements Semantic Data Elements provide interoperability Semantic Data Elements are used by:

Processes authored in the Process authoring environment (Protean)

Proteus engine Greed and the Rules Engine to understand what

Proteus Engine is expecting inferences for

Page 25: Semantic Data Capture Initiative

Knowledge Development Processes Selected

Follow-up visit for Hypertension Patients Uncomplicated Upper Respiratory Tract Infection

Selected Processes were analyzed

Improvement Opportunities were identified in the Processes

Proteus Process (knowledge) Authoring

Ongoing Iterative refinement

Deployment of Proteus processes

Page 26: Semantic Data Capture Initiative

Interaction between CPNG and SDCI Apps – Metadata API

To be used only at author time

Method suggestions searchDataElements() getDataElementByValue() getDataElementsByFormName() getDataElementsByFormID() getDataElementsByModuleName() getDataElementsByModuleID()

Page 27: Semantic Data Capture Initiative

CPNG Web App

SDCI Web App

CPNG Web Service

hyperlink with the encrypted authentication

info

Method call using encrypted

authentication info

Interaction between CPNG and SDCI Apps – Patient Data API

For accessing existing patient data, during run-time Aim is to prevent re-entry of data already available in

CPNG Save data collected during execution of a Proteus

process to be available for future use by CPNG or SDCI App users

Page 28: Semantic Data Capture Initiative

Interaction between CPNG and SDCI Apps – Patient Data API

Suggested Methods getAllPatientByDataValues() for the time range getPatientDataValue() for the data element specified by

data element id savePatientData() for the specified patient and given

data element id

Page 29: Semantic Data Capture Initiative

thank you

Page 30: Semantic Data Capture Initiative

Greed(Rules Authoring Tool)

Protean(Process Authoring Tool)

CPNG (EMR)

MetadataPatient

Data

vMRvMR

Patient Data

TemplateTemplate

Henry Ford KR

KC

Proteus Engine(Process/Guideline Engine)

A B

D

C

E

C1

Rules Engine

Morningside KR

KC

The Big Picture

Page 31: Semantic Data Capture Initiative

Deployable Guideline

A B

D

C

E

C1

Value of Editability – Effort and Time Reduction

DevelopersDevelopersDomain ExpertsDomain Experts

Consensus Building Initial Development

Testing

Modification

Without Authoring Tool

Page 32: Semantic Data Capture Initiative

Deployable Guideline

A B

D

C

E

C1

Value of Editability – Effort and Time Reduction

Domain ExpertsDomain Experts

Consensus Building

Testing

Authoring

With Authoring Tool

Page 33: Semantic Data Capture Initiative

Value of Editability – Applicability to different Locales

DiseaseDisease PopulationPopulation Medical Setup/Skills

Medical Setup/Skills

Level A

Level A Guideline

Level A Guideline

Top Level

Top - Level Guideline

Top - Level Guideline

Level B

Level B Guideline

Level B Guideline

Page 34: Semantic Data Capture Initiative

Evaluation


Recommended