EpiS3: a semantically interoperable social network for syndromic surveillance and disease control

1
EpiS3 a semantically interoperable social
network for syndromic surveillance and disease
control

• Luciana Tricai Cavalini and Timothy Wayne Cook
• National Institute of Science and Technology
  Medicine Assisted by Scientific Computing

2
Summary

• The problem
• The current solution
• Remaining challenges
• A new approach
• Implementation
• Future steps

3
The problem

• Syndromic Surveillance

4
Problem 1 Detecting Cases
First cases detected
Index case
5
Fever? Bleeding? Jaundice?
6
Problem 2 Decision Making
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
The current solution

• Syndromic Surveillance

11
Current solution Standardize the data model
12
(No Transcript)
13
(No Transcript)
14
The Current Solution Issues

• Top-down data models
• Risk of inaccurate or incomplete data
• Hospital/clinic centered applications
• No records from uncovered populations
• Incipient Decision Support Systems (DSS)
• Mostly academic projects in internal medicine

15
Remaining issues

• Syndromic Surveillance

16
Problem is Accuracy or Utility?
17
(No Transcript)
18
Remaining Questions

• How to collect data in the most opportune moment?
• At the point of care
• In the household
• How to get data with proper
• ...accuracy...
• ...granularity...
• ...that will allow implementation of useful DSS
  for syndromic surveillance?

19
How to get...
Dr. Cool
Your patient Jane
Updated her problem list on Apr 29, 2014 533pm -
Fever YES - Bleeding YES - Location Nose
Suspicious case of Acute Febrile Hemorrhagic
Syndrome
What to do
...without creating another data silo?
20
A new approach

• Syndromic Surveillance

21
Fever? Bleeding? Jaundice?
22
MedWeb 3.0 Plugin Suite
Rabies prophylaxis app
Hospital infection control app
AFJHS app
And so on
Bioterrorism app
Poisonous animals app
Minimalistic, XML-based MMD technology
MLHIM-based implementation
Multilevel Model-Driven Approach
Harmonization
23
implementation

• Epidemiological Surveillance Support System
  (EpiS3)

24
Acute Febrile Jaundice Hemorrhagic Syndrome
(AFJHS) App
gt 1 y/o Fever 0-3 wks Jaundice

• gt 1 y/o
• Fever 0-3 wks
• Bleeding signs

• gt 1 y/o
• Fever 0-3 wks
• Jaundice and Bleeding

AFJS
AFHS
AFJHS
Treat malaria
Malaria blood smear test
Positive
Negative
Evaluate current epidemiological profile of the
territory
Hepatitis Yellow Fever Leptospirosis Sepsis Typh
oid Fever

• Dengue
• Sepsis
• Meningococcemia
• Typhoid Fever
• Hantavirus
• Other Arbovirosis

Hepatitis Yellow Fever Leptospirosis Sepsis Typh
oid Fever
AFJHS
AFJS
AFHS
25
Reference Model
Concept
Constraint
Definition
26
Concept Constraint Definition Generator (CCD-Gen)
www.ccdgen.com
27
CCD Library on CCD-Gen
www.ccdgen.com/ccdlib
28
AFJHS App Form on CCD-Gen
29
AFJHS App CCD Schema
30
AFJHS App Sample Data Instances
AFHS with spontaneous bleeding
31
AFJHS App Sample Data Instances
AFHS with tourniquet test positive
32
AFJHS App Sample Data Instances
AFJS with mucosa jaundice
33
Already Implemented
16 AFJHS simulated cases (all possible
classifications)
AFHS
AFJHS
AFJS

• Spontaneous mucosa
• Spontaneous skin
• Spontaneous both
• Tourniquet mucosa
• Tourniquet skin
• Tourniquet both
• Malaria

• Spontaneous bleeding
• Tourniquet test

• Mucosa
• Skin
• Both

• Age
• Fever
• Fever duration
• No signs

Negative
a R library that converts the XML data
instances into R data frames
34
(No Transcript)
35
Future steps

• Epidemiological Surveillance Support System
  (EpiS3)

36
EpiS3 Future Steps

• App User Interface
• Desktop and mHealth versions
• DSS Algorithms
• Clinical evaluation
• Messaging
• Reporting
• EpiInfo Form Builder for MLHIM data

37

• Thank you!
• lutricav_at_lampada.uerj.br
• tim_at_mlhim.org
• google.com/MedWeb30