Transcript mekala-presentation-notifier-0103

NOTIFIER
DREAMS
Vijaya Mekala
The University of Texas Health Science Center at Houston
Arden Syntax SIG
Jan 14, 2003, San Antonio, TX.
1
Decision Support Roles and Messages
Input Generating Module
User
Database
Notification
System
Decision Support
System Roles
Inter-role Message
DREAMS
Event
Monitor
2
Inference Recipient
Engine
Example:
User updates patient data
with an associated rule, event
monitor detects event (change
in database), calls inference
engine, rule generates output
and sends it to notifier which
routes output to proper
recipient
Notifier / Notification Overview
Input Generating Module
Notifier
DREAMS
Gateways
(fax, e-mail, pager, etc.)
MPI
3
Destinations
Destinations
A Notifier System
Key Terms

Input Generating Module
– An application/CDSS that generates the XML formatted clinical
message

Notifier
– A programming framework that processes, generates and routes
personalized and timely notifications to the user via a range of
pathways subject to user preference.
– Used for on-ambulance CDSS
– Used to alert physicians to request for assistance
DREAMS

Triggering message
– The XML formatted clinical message which the input generating
module triggers

Notification
– The formatted message appearing to the end user on the device of his
choice.
4
A Notifier System
A Bird’s eye view
IGM
Notifier
Triggering Message
DREAMS
The Input Generating Module (CDSS)
Triggers XML Clinical Message
 XML Message routed to Notifier
Notifier
Receive XML message
Process XML message as per pre-determined BI
Format Notification as per end-user device
Destinations
Display Notifications
5
Destinations
Notification
Notifier
A Schematic view
Destinations
Notifier
CDSS
Triggering
Message
Notification
DREAMS
Triggering
message
Message
Collector
6
Notification
Message
Processor
Message
Distributor
Notifier
Schematic view (cont….)
 Message Collector
Store the collected input messages in the Database
DREAMS
 Input Message received as XML Document
 XML Document Loaded via XML API
 API converts Document into Database-Friendly Format
XML document
DSS
7
XML API
Collector Table
Notifier
Schematic view (cont….)
 Message Processor:
Process the Stored Message subject to Business Logic
DREAMS
 Extract Data in Collector Table and ensure Business Logic
 Business Logic is ensured by using JAVA/COM components
 Notification is generated and stored in the Database
 Notification is passed on to distributor for further delivery
 System also ensures Fail Safe delivery
 Fail safe delivery is established by escalating messages sequentially and conditionally
Process the data according
to the rules (Java or COM)
and then writes the
notifications
Message collector
table
8
Notification
Notification table
Notifier
Schematic view (cont….)
 Message Distributor:
Format and Distribute Notification to end-user
 Receive Stored Notification from Processor
 Achieve Device-Specific format using XSLT
 Deliver Notification using various Protocols (SMTP, HTTP etc.)
DREAMS
Formatter
XSLT
Notification Table
9
HTTP
Application Architecture


DREAMS

10
Data tier:
Storage of the events
and notifications in a
database for persistent
record information.
Business Tier:
Business objects and
rules
Data Manipulation and
transformation into
information that could
be used in a stateful
manner
Presentation tier:
Formatting the
notification into a
device specific format.
Data Logic
Oracle
SQL Server
Oracle
SQL Server
Business Logic
Presentation Logic
Java/COM components
XSLT
Java
COM
Components
Xpath / XSLT
Deployment Architecture


Configures where the subsystems are
located and how they operate.
Notifier can be deployed on a single
server or on multiple servers.
MC
MP
Database
MC - Input message Collector
MP - Message processor
MD - Notification Distributor
MD
Single server
DREAMS
Server
Pros:
Single Server
Communication
Compatibility
Cons:
Failure
MC
Multiple Server
MP
Pros:
Scalability
Failure
Availability
Cons:
Communication
Compatibility
11
Database
MD
Server
Server
Multiple Server
Questions
DREAMS
12
????