Transcript Designing A Patient Monitoring System Using Cloud And Semantic

Technical University of Crete
Designing A Patient
Monitoring System
Using Cloud And Semantic
Web Technologies
Chryssa Thermolia -Ekaterini S. Bei
Stelios Sotiriadis - Kostas Stravoskoufos
Euripides G.M. Petrakis
17th International Conference on Brain and Health Informatics (ICBHI 2015)
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"
Motivation
Patient Monitoring Systems


New era of healthcare requires new tools, devices and systems to improve
health
services​
Advanced health care increases the need for constant monitoring of patient's
condition​, especially in chronic diseases
​Solution

Multi-source patient monitoring evolves into an important service in this
domain​

Patient monitoring systems offer advantages in
- early-diagnosis
- optimal treatment strategies
- disease prevention​
- analysis, management and communication of medical information​
Key factors in this attempt
o
integration of medical information from various sources
o
constant, on-time briefing of patient’s health state and behavior
17th International Conference on Brain and Health Informatics (ICBHI 2015)
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"
Background
Semantic Web
Semantic Web: Collection of standard technologies to realize a Web of
Data
- Ontology: Heart of Semantic Web
Formal representation of knowledge as a set of concepts.
Describes the concepts (classes) of in a domain interest, their
characteristics (data properties) and the relationships that hold
between the concepts (object properties)

Example:
o
o
OWL (Web Ontology Language):
concepts and their relationships.
SWRL (Semantic
reasoning in OWL
Web
Rule
Language that describes the
Language):
Implements
deductive
17th International Conference on Brain and Health Informatics (ICBHI 2015)
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"
Background
Internet of Things (IoT)

Internet of Things (IoT)
Relates users and their smart devices
along with sensors used in every day
actions
(e.g., Smart phones and wearable
devices)
Various devices could offer to sensor
embedded healthcare new applications
and services


IoT is expected to greatly
transform
the
healthcare
industry by improving the
clinician-patient relationship
Clinicians using IoT could
- monitor patients remotely
- run a diagnosis in real-time
- be notified for sudden and
abnormal events and act
immediately
17th International Conference on Brain and Health Informatics (ICBHI 2015)
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"


Background
IoT growth leads to large amounts of
data
Need for big data storage, processing
and
accessing
Cloud computing as a paradigm for big
data storage and analytics
Cloud Computing
Cloud Computing:

Provides a platform environment where
hardware and software could be
delivered on a bespoke manner to users
and utilized accordingly to their
requests.
Allows the scaling of user resources on
demand (elasticity)

Combination of IoT and Cloud
Computing is the real innovation
17th International Conference on Brain and Health Informatics (ICBHI 2015)
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"
Core
System
The designed patient monitoring
system is aimed to be able to
support clinical professionals during
the initial evaluation and diagnosis of
adults with suspected BD, and during
their treatment
The system will be able to provide
 evidence-based treatment options
for a personalized therapeutic
approach
 notifications for early-warning
signs and alerts for crucial mood
swings
Our design consists
components
of
three
1) The implemented core system.
2) A proposal of the front-end system.
3) The vision of the back-end system.
17th International Conference on Brain and Health Informatics (ICBHI 2015)
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"
Core System
Guidelines





The World Federation of Societies of Biological Psychiatry (WFSBP) Guidelines for the
Biological Treatment of Bipolar Disorders (long-term treatment)
Canadian Network for Mood and Anxiety Treatments (CANMAT) and International
Society for Bipolar Disorders (ISBD) collaborative update of CANMAT guidelines for the
management of patients with bipolar disorder (acute episodes of mania, depression)
The CANMAT task force recommendations for the management of patients with mood
disorders and comorbid medical conditions (diagnosis)
Australian and New Zealand clinical practice guidelines for the treatment of bipolar
disorder (breakthrough depression)
Bipolar disorder algorithms: The Psychopharmacology Algorithm Project at the Harvard
South Shore Psychiatry Program , TEXAS Medication Algorithm Project (immediate &
urgent)
Rating Scales
o
o
o
o
o
o
Hamilton Depression Rating Scale (HDRS)
Young Mania Rating Scale (YMRS)
Montgomery–Åsberg
Clinical Global Impression Bipolar Version Scale, CGI-BP
Global Assessment of Functioning Scale, GAF
The Quality of Life in Bipolar Disorder Questionnaire
17th International Conference on Brain and Health Informatics (ICBHI 2015)
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"
Core System
User Scenarios For BD
 Types of Bipolar Disorder:
 Diagnosis:
DSM-IV-TR Classification
1st Level (clinician studies the
patient’s experience in regards of
abnormal symptoms),
2nd Level (clinician estimates
according to defined criteria taking
also into account family history)
 Therapy: pharmacotherapy
(mood stabilizers, antidepressants,
antipsychotics), psychoeducation,
psychotherapy
Bipolar I Disorder (BDI): one or more
episode of mania with or without
major depressive episodes
Bipolar II Disorder (BDII): one or more
episode of hypomania as well as at least
one major depressive episode with no
psychotic features
Cyclothymic disorder: low grade cycling of
mood with the presence or history of
hypomanic episodes and periods of
depression that do not meet criteria for
major depressive episodes
Bipolar disorder NOS: Bipolar symptoms that
do not meet the criteria for previous
subtypes
17th International Conference on Brain and Health Informatics (ICBHI 2015)
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"
Core System
User Scenarios For BD
Possible phase transitions of
Bipolar I Disorder
The
scenarios
are
developed considering
possible
phase
transitions of BD that
may occur during the
progress of the disease
(mania
to
euthymia,
depression to euthymia,
mania to depression, and
vice versa)
17th International Conference on Brain and Health Informatics (ICBHI 2015)
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"
Core System
User Scenarios
The developed scenarios support clinician through

Diagnosis

Immediate & Urgent Management & Treatment

Acute Manic Episode Treatment

Acute Depressive Episode Treatment
•

Breakthrough Depressive Episode (Li) Treatment
Long-term Treatment
17th International Conference on Brain and Health Informatics (ICBHI 2015)
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"
Core System
Ontology
Dynamic Entities






PHR
PatientState
Symptom
Function Tests
Therapy
Medicine
Static entities




Patient
PatientHistory
Episode
InitialEvaluation
•
History
•
Questionnaire
(MDQ, BSDS,
CIDI)
•
Clinical
Evaluation
 Medical Cause
 Diagnosis
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"
Clinical guidelines encoded as SWRL
rules to issue alerts and
recommendations
Apply rules over patient’s information.

Core System
SWRL Rules
Example:
If there is a positive mood
questionnaire, there is a suspicion of
BD and in that case, if the clinical
evaluation excludes other medical
causes from being responsible for the
patient’s symptoms then, the rule
concludes that the clinician needs to
continue with assiduous clinical
examination
PHR ∩ InitialEvaluation ∩ (∃
Questionnaire.result = true) ∩ (∃
ClinicalEvaluation.result = normal) →
Recommendation
17th International Conference on Brain and Health Informatics (ICBHI 2015)
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"
Overall Architecture
High level expected
functionalities of the
monitoring
system
prototype:

Data collection






Interoperability
Notification
services
Data analysis and
integration
Secure data
storage
Legacy system
adaptors
Other services
such as semantic
analysis tools
17th International Conference on Brain and Health Informatics (ICBHI 2015)
"Designing A Patient Monitoring System Using Cloud And Semantic Web Technologies"
Conclusions &
Future Work
Conclusions:


Analyze patient records and filtering
evidence-based guidelines to offer
individualized notifications and
recommendations for diagnosis
and treatment.
Propose a cloud deployment model
as a perspective for an advanced
environment that assists in
monitoring of complex chronic
pathologies, such as brain disorders
including BD.
Future Work:

Implement and test the cloudbased
architecture on a real setting
performing data acquisition from sensors
and wearable devices
17th International Conference on Brain and Health Informatics (ICBHI 2015)
Thank you!
(Questions ?)