Transcript Assert facts - UMBC ebiquity research group

Context Aware Surgical
Training Environment
Operating Room of the Future
23 June 2006
Vision
• ORs and surgical training centers will be pervasive
computing environments
• Devices, sensors, tags, trainers, PDAs, monitors will
discover one another and interoperate
• Components will require access to a context model
to manage resources effectively
• The context model includes relevant information on people,
roles, activities, events, workflow, devices, …
• Intelligent components will be able to recognize
events and activities
• Even in the presence of noisy or incomplete data
SimCenter Context Model
• The people, devices, environmental factors,
information resources, …
• The activities and events going on and the roles
people are playing in them
• The anticipated schedules, workflow models, training
plans, …
• Profiles of the training resources – capabilities,
complexity, ...
• Profiles of the trainees, their background, training
goals, training history, training plans
• Privacy and information sharing constraints and
policies
SimCenter Interoperability
• Interoperability means that devices and systems can
share and use models, data and services as needed
• Training resources can be intelligently matched to
the training needs and profiles of trainees
• Constructing and monitoring customized training
experiences, enabled by data and model
interoperability
• Capability of end-to-end simulation fly throughs
ORs will be data rich
Drugs
Tools
CAST
Patient Monitors
Staff
• ORs will be awash in low-level data, much of it noisy or
incomplete
• Challenges include coping with the noise and interpreting the
low-level data to recognize high-level events and activities
Streaming Database Engines
TelegraphCQ Data Stream Management System from UC Berkeley
Queries
Index
Result
Query
Result
Index
Data
Traditional DBMS
• Data is stored/indexed in system
• Queries applied to stored data as
they “stream through”
Stream Management System
• Queries stored/indexed in system
• Data applied to stored queries as
they “stream through”
Example Streams & Queries
• Data Sources (DS)
• Continually send data to streams
• PUSH or PULL sources
• Stream
create stream traffic.incidents (
incidentID
integer,
city
varchar(100),
street
varchar(100),
description
varchar(1000),
tcqtime
timestamp TIMESTAMPCOLUMN
) type archived;
2207, Dublin, JWO HACIENDA DR, Traffic Hazard - Debris/Objects, Wed Jul 16 18:37:00 PDT 2003
• Continuous queries must be specified over a time window
• Select count(*) from mer.rfid [range by ‘30 seconds’ slide by ‘10
seconds’] group by tag_id
• Window is relative to the most recently arrived tuple.
System Architecture
Continuous
Queries
Trend
Analyzer
Patient Monitor
Medicines
Tools
Stream
Processor
(TelegraphCQ)
Context
Aware
Agent
Rule
Base
RFID
System
Video
Clipper
Database
Staff
Patient
History
Staff
Medical
Supplies
Medical
Encounter
Record
Event Detection - Level 3
Video
Clipper
Medical
Encounter
Record
Rule
Base
Event Detection - Level 2
Staff
Patient
History
Database
Medical
Supplies
Low-Level
Event
Processor
Trend
Analyzer
Event Detection - Level 1
Stream
Processor
(TelegraphCQ)
RFID
System
Patient Monitor
Continuous
Queries
Medicines
Tools
Staff
Simulations and Results
•
•
•
•
•
The Human Patient Simulator (HPS) from METI
Designed to react like a human
Used for training resident doctors
Responds to medical treatment
Physiological data sets from HPS
Scenario and Patient Profile
• HPS can run patient profiles
• Data logs from simulations used
to evaluate the system
• Significant events for a blunt
trauma multiple injuries profile
include hypovolemia, tension
pneumothorax, decompression
and fluid infusions
• Provides data for Medical
Encounter Record
• Ran 30 simulations on 7 profiles
measuring false positives &
negatives and latency in
detecting events
Patient Profile
Status and Plans
• We’re building on work done for Trauma Pod
• Preliminary work has produced a scalable
architecture based on TelegraphCQ, the Jess rule
based reasoning engine and relational databases
• Fall 2006:
• Install equipment in SimCenter in Fall 2006
• Investigate use of LiveData’s system for output
• Develop ontologies in RDF/OWL to support semantic
interoperability
• Spring 2007
• Develop and implement scenarios for proof of concept and
evaluation
Key Enablers
• Semantic interoperability
• Develop and use standard ontologies and data models to
maximize information sharing
• Use evolving standards (XML, RDF, …)
• Greater context awareness and intelligence
• The whole is more than the sum of its parts -- drawing
inferences, integrating across all entities, and recognizing
patterns
• A service oriented architecture
• Distributed programs and resources share common
interface standards and publish their APIs
http://ebiquity.umbc.edu/