Towards a Cost-Effective Homecare for
a Caregiver Assistance System in Brazil
MAURO OLIVEIRA
LAR-A
Computer Network Laboratory of Aracati
Slide 1
Team working on this project
Towards a Cost-Effective Homecare for
a Caregiver Assistance System In Brazil
Mauro Oliveira
Odorico Andrade
Marcos Santos
Roberto Alcântara
Federal Institute of Ceará (IFCE)
Aracati, Brazil
[email protected]
Federal University of Ceara (UFC)
Fortaleza, Brazil
[email protected]
State University of Ceara (UECE)
Fortaleza, Brazil
[email protected]
Federal Institute of Ceará (IFCE)
Aracati, Brazil
[email protected]
Germanno Teles
Nazim Agoulmine
Northeast Bank of Brazil(BNB)
Fortaleza, Brazil
[email protected]
Joseph Fourier University (UJF)
Université d´Evry Val dÉssone
[email protected]
16th
Healthcom
October 16th, 2014
Natal - RN
Slide 2
Summary
1. Contextualization
2. Application Scenario
3. Brazilian Digital TV
4. LARIISA Project
5. Prototype
Conclusion
Slide 3
1. Contextualization
Slide 4
CONTEXTUALIZATION
Health System - Information Era
Based on Disease PREVENTION
Encouraged
Costs
¢
Primary
Health Care
Descentralization of Public
Health System
Health Agent
$
Discouraged
Hospital (specialits)
Fonte: K. Jennings, K. Miller, S. Materna (1997)
Slide 5
CONTEXTUALIZATION 
PROBLEM
Message
Management
Increasing Complexity of
Health Management for
Decision-making
Information
Data
Acquisition
Slide 6
CONTEXTUALIZATION  PROBLEM 
SOLUTION
CONTEXT-AWARE FRAMEWORK
DecisionMaking
Inference
Mechanism
ONTOLOGY
BAYESIAN
NETWORK
DATA MINING
Information
LARIISA: an Intelligent System to
support decision-making process
Health
Knowledge
Geolocation
Metadata
Slide 7
LARIISA’s Scenario: a context-aware framework
ONTOLOGY
BAYESIAN NETWORKS
DATA MINING
Data Acquisition
(Patient CONTEXT)
METADATA
Option 1
LARIISA:
A Context-Aware Framework
Option 2
Option 3
DECISION-MAKING
Slide 8
2. Application Scenario
Slide 9
Scenario for LARIISA Application
CAREGIVER - an unpaid or paid person who helps another individual
with an impairment with his or her activities of daily living.
LARIISA can help
Specialized person
2. Motivation:
The Brazilian Digital TV
Or NON specialized person
Slide 10
Data acquisition - CONTEXT
Prague, Czech Republic
July/2013
Slide 11
LARIISA Data Acquisition
Slide 12
LARIISA: Next Generation
Data Acquisition
(CONTEXT)
ONTOLOGY
BAYESIAN NETWORKS
DARA MINING
METADATA
DECISION-MAKING
Slide 13
3. The Brazilian Digital TV
Slide 14
The Brazilian Digital TV
Today
Interactive Digital TV
Analógico
Slide 15
1. MOTIVATION
Interactive
Digital TV
Audio
Video
Data
Data Carrossel
Network
Slide 16
The Brazilian Digital TV
Interactive Digital TV
Slide 17
Architecture of the Brazilian Digital TV
GINGA, a recommendation H.761 of the International Telecommunications Union (ITU-T).
Prague, Czech Republic
July/2013
Slide 18
4. LARIISA Project
Slide 19
LARIISA: a Context-Aware Framework
Atmospheric
Global
Positioning
Geographical
Humidity
Sensor
Temperature
Sensor
Internet
Accelerometer
Connection
Compass
Digital
Camera
Light
Sensor
Proximity
Gyroscope
Sensor
Information
System
Pressure
SystemSensor
(GPS)(GIS)
Medical Sensors
Slide 20
LARIISA: a Context-Aware Framework
Patient A
Patient B
Information: lat=S 2° 22' 32.9483“, lon=W 34° 33' 21.4657“ on March
24th, 2013 at 3:00PM. Body temperature=37°C, Heart rate=90 bpm,
Blood pressure=120/80. Address= Av. da Sé, n° 227, Conj. Palmeiras,
Fortaleza – CE. Local Temperature=32°C. Symptoms=Chills, Diarrhea.
Information: lat=S 3° 45' 48.6429“, lon=W 38° 36' 28.7434“ on March
3rd, 2013 at 5:00PM. Body temperature=40°C, Heart rate=110 bpm,
Blood pressure=140/90. Address=Av. F, 126-298-Conj. Ceará, Fortaleza
– CE. Local Temperature=25°C. Symptoms=Headache, Vomiting, Body
aches.
Who is the patient?
Are
Data
Structured?
80%-90%
likelihood
of being with
Dengue (Local Context)
The photo IMG001 was taken at lat=S 3° 45' 48.6532“,
lon=W 38° 36' 28.7332“ on February 2nd, 2013 at
8:00AM. Address =Av. G, 162-229-Conj. Ceará,
Fortaleza – CE. Local Temperatura=29°C. Comment=
Dengue Habitat.
Health Agent
Relocating a health agent for the Patient B’s house
15th
Healthcom
October 10th, 2013
Lisbon, Portugal
Slide 21
Building a
metadata
file
<lat=S 3° 45' 48.6429“>, <lon=W 38° 36' 28.7434“>, <time=17:00h>,
<date=03/02/2013>, <body_temp=40°C>, <heart_rate=110 bpm>,
<blood_press=140/90>, <loc_add=Av. F, 126-298-Conj. Ceará,
Fortaleza – CE>, <loc_weather=25°C>, <symptom=A, B, C>,
<sus_id=209968974640021>
<lat>3° 45' 48.6429"</lat>
<lon>38° 36' 28.7434"</lon>
Geolocation
<time>17:00</time>
<date>03/02/2013</date>
<body_temp>40°C</body_temp>
Health
Information
metadata file
<heart_rate>110bpm</heart_rate>
<blood_pressure>140/90</blood_pressure>
<loc_name>Av. F, 126-298-Conj. Ceará, Fortaleza - CE</loc_name>
<symptom>A, B, C</symptom>
Patient Identification
<sus_id>209968974640021</sus_id>
Slide 22
LARIISA’s Architecture: a context-aware framework
Data Acquisition
Context Providers
Internet
User
Device
Health Agent
Device
Data Processing
Publishing
Symptoms + sus_id
Metadata
Security Protocol
Context Aggregator (CA)
Inference
Rules
Global
Context
Local
Context
Health Managers
System
Slide 23
LARIISA: Functional Diagram
INFERENCE MODULE
SITUATION ROOM
IN
User Interface
LARIISA_BAY
Inference Module
Specialist Decision
Module
%
Specialist
2
Specialist
Validation:
A f(%)
A ≠ A’
3
Sensors
Pass
Through
A f(%)
A = A’
METADATA
A’
Patient
Specialist
C’
Specialist
Decision:
f(%)
RB
OUT
B’
1
Patient
Health
Agent
Decision
Module
Interface
A
A’
B
B’
C
C’
A
A’
B
B’
C
C’
Health
Agent
Manager
Health Center
Ambulance
Epidemic Graph
OTHER
CONTEXT
PROVIDERS
Inference
Rules
Global Context
Repository
Local Context
Repository
LARIISA
Slide 24
5. Prototype
Slide 25
LARIISA’s Prototype
ONTOLOGY
BAYESIAN NETWORKS
Data Acquisition
(Patient CONTEXT)
METADATA
Option 1
LARIISA:
A Context-Aware Framework
Option 2
Option 3
DECISION-MAKING
Slide 26
Prototype: Dengue Fever Case Study
Local health context model
Metadata
Global health context model
Slide 27
Screens of the
ENTRADA
proposed
System
DO
SISTEMA
Interface
Módulo de
Decisão
Interface do Usuário
1
Decisão do
Especialista:
f(%)
Paciente
RB
Patient
Health Agent
Especialista
Specialist
Sensores
Módulo de
Decisão
A’
SAÍDA DO
SISTEMA
B’
Módulo de
Inferência do
LARIISA_Bay
Agente de
Saúde
SALA DE SITUAÇÃO
2
%
Validação do
Especialista:
A f(%)
A ≠ A’
3
Pass
Through
A f(%)
A = A’
C’
A
Paciente
Especialista
Agente de
Saúde
Gestor
A’
B
B’
C
C’
A
A’
B
B’
C
C’
Posto de Saúde
Ambulância
Gráfico de Epidemias
METADADO
OUTROS
PROVEDORES
DE CONTEXTO
LARIISA_Bay
Regras de Repositório de Repositório de
Inferência Contexto Global Contexto Local
LARIISA
Slide 28
Slide 29
Conclusion
Slide 30
Sponsors
LARIISA project is being sponsored since 2004 by
the Science and Technology Ministry of Brazil and others
Brazilian Research Agencies
It will be applied to the brazilian public health system
Diga Saude (FUNCAP)
SISA (FIOCRUZ)
LARIISA (IFCE)
GISSA (FINEP)
Next Saude (DATASUS / MinConclusão
Saúde)
Slide 31
Mauro Oliveira
[email protected]
+55 85 9705 4321
www.maurooliveira.com.br
15th
Healthcom
October 10th, 2013
Lisbon, Portugal
Slide 32
MUITO OBRIGADO
Slide 33