Enabling Personalized Composition and Adaptive Provisioning of Web Services
Smart Data enabling Personalized Digital Health
-
Upload
amit-sheth -
Category
Education
-
view
3.673 -
download
1
description
Transcript of Smart Data enabling Personalized Digital Health
Smart Data enabling Personalized Digital Health: Deriving Value via harnessing Volume, Variety and Velocity
using semantics and Semantic Web
Put Knoesis Banner
Pramod Anantharam
Amit P. Sheth
Cory Henson
Dr. T.K. Prasad
Contributions by many, but Special Thanks to:
Ohio Center of Excellence in Knowledge-enabled Computing (Kno.e.sis) Wright State University, USA
Sujan Perera
Delroy Cameron
2
A Historical Perspective on Collecting Health Observations
Diseases treated onlyby external observations
First peek beyond justexternal observations
Information overload!
Doctors relied only on external observations
Stethoscope was the first instrument to go beyond just external
observations
Though the stethoscope has survived, it is only one among many observations
in modern medicine
http://en.wikipedia.org/wiki/Timeline_of_medicine_and_medical_technology
2600 BC ~1815 Today
Imhotep
Laennec’s stethoscope
Image Credit: British Museum
3
The Patient of the FutureMIT Technology Review, 2012
http://www.technologyreview.com/featuredstory/426968/the-patient-of-the-future/
4
Big Data in Medicine: Implications
“We should not make the mistake of seeing data as a technical issue. It’s a synthesis problem. That’s because information is not the scarce
resource. Attention is.”-- Conrad Wai, The data addiction | The Ideas Economy
http://www.davidscaduto.com/post/9048831674/we-should-not-make-the-mistake-of-seeing-data-as
5
Sources of Big Data in Digital Health
Velocity Volume
Variety
Veracity
Image: http://www.dr4ward.com/dr4ward/2013/04/what-is-the-power-of-the-big-data-in-healthcare-infographic.html
6
Future Interoperability Challenges: 360 degree health
7
Big Data in Digital Health: Can alerts work?
"According to multiple recent studies, doctors ignore between 49–96% of all CDS alerts that EMRs give them.”1
"Clinical Decision Support systems link health observations with health knowledge to influence health choices by clinicians for improved health care".
-- Robert Hayward, Centre for Health Evidence
1http://www.fastcodesign.com/1664763/badly-designed-electronic-medical-records-can-kill-you
8
Information Overload leading to Alert Fatigue
Ignoring alerts is not limited to Emergency Rooms but has also crept into EMR alerts commonly referred to as “alert fatigue”
http://health.embs.org/editorial-blog/noise-in-hospital-intensive-care-units-icus/
9
• What if your data volume gets so large and varied you don't know how to deal with it?
• Do you store all your data?• Do you analyze it all?• How can you find out which data points are
really important?• How can you use it to your best advantage?
Questions typically asked on Big Data
http://www.sas.com/big-data/
10http://techcrunch.com/2012/10/27/big-data-right-now-five-trendy-open-source-technologies/
Variety of Data Analytics Enablers
11
• Prediction of the spread of flu in real time during H1N1 2009– Google tested a mammoth of 450 million different mathematical
models to test the search terms, comparing their predictions against the actual flu cases; 45 important parameters were founds
– Model was tested when H1N1 crisis struck in 2009 and gave more meaningful and valuable real time information than any public health official system [Big Data, Viktor Mayer-Schonberger and Kenneth Cukier, 2013]
• FareCast: predict the direction of air fares over different routes [Big Data, Viktor Mayer-Schonberger and Kenneth Cukier, 2013]
• NY city manholes problem [ICML Discussion, 2012]
Illustrative Big Data Applications
12
• Current focus mainly to serve business intelligence and targeted analytics needs, not to serve complex individual and collective human needs (e.g., empower human in health, fitness and well-being; better disaster coordination, smart energy consumption) that is highly personalized/individualized/contextualized– Incorporate real-world complexity: multi-modal and multi-sensory nature of real-
world and human perception– Need deeper understanding of data and its role to information (e.g., skew,
coverage) – Beyond correlation -> causation :: actionable info, decisions grounded on insights
• Human involvement and guidance: Leading to actionable information, understanding and insight right in the context of human activities– Bottom-up & Top-down processing: Infusion of models and background knowledge
(data + knowledge + reasoning)
What is missing?
13
Contextual
Information Smart Data
Makes Sense
Actionable or help decision support/making
14
DescriptiveExploratoryInferentialPredictive
Causal
Improved Analytics CREATION
PROCESSING
EXPERIENCE & DECISION MAKING
Human Centric Computing
15
Smart Data
Smart data makes sense out of Big data
It provides value from harnessing the challenges posed by volume, velocity, variety and veracity of big data, in-
turn providing actionable information and improve decision
making.
16
“OF human, BY human and FOR human”
Smart data is focused on the actionable value achieved by human
involvement in data creation, processing and consumption phases
for improving the human experience.
Another perspective on Smart Data
17
• Focus on verticals: advertising‚ social media‚ retail‚ financial services‚ telecom‚ and healthcare
– Aggregate data, focused on transactions, limited integration (limited complexity), analytics to find (simple) patterns
– Emphasis on technologies to handle volume/scale, and to lesser extent velocity: Hadoop, NoSQL,MPP warehouse ….
– Full faith in the power of data (no hypothesis), bottom up analysis
Current Focus on Big Data
18
“OF human, BY human and FOR human”
Another perspective on Smart Data
19Petabytes of Physical(sensory)-Cyber-Social Data everyday!
More on PCS Computing: http://wiki.knoesis.org/index.php/PCS
‘OF human’ : Relevant Real-time Data Streams for Human Experience
20
“OF human, BY human and FOR human”
Another perspective on Smart Data
Use of Prior Human-created Knowledge Models
21
‘BY human’: Involving Crowd Intelligence in data processing workflows
Crowdsourcing and Domain-expert guided Machine Learning Modeling
22
“OF human, BY human and FOR human”
Another perspective on Smart Data
23
Weather Application
‘FOR human’ : Improving Human Experience
Detection of events, such as wheezing sound, indoor
temperature, humidity, dust, and CO2 level
Weather ApplicationAsthma Healthcare Application
Action in the Physical World
Close the window at home during day to avoid CO2 inflow,
to avoid asthma attacks at night
Public Health
Personal
Population Level
24
Why do we care about Smart Data rather than Big Data?
25
Smart Data enabling Personalized Digital Health: Deriving Value via harnessing Volume, Variety and Velocity
Amit P. Sheth, Kno.e.sis, Wright State University
26
April 6, 2011
http://www.nbcnews.com/id/44599555/#.UmMHMWRDszQ
Mr. Michael Yocabet suffering from type 1 diabetes is recommended a kidney transplant at the University of Pittsburgh Medical Center. The
organ donor is his life partner Ms. Christina Mecannic
27
May 6, 2011
http://www.nbcnews.com/id/44599555/#.UmMHMWRDszQ
The couple leaned about the botched kidney transplant making the situation of Mr. Yocabet much worse! The kidney he got from his wife
has infected him with Hepatitis C aggravating his health issues.
28
Life Threatening Implications!
http://www.scientificamerican.com/article.cfm?id=2003-blackout-five-years-later
Mr. Yocabet was a disabled former truck driver and he has diabetes type 1. Treatment for the liver may harm his kidney even cause organ failure and death!
“Because he’s on anti-rejection drugs, the hepatitis C will be a lot worse in him,” -- Ms. Christina Mecannic
29
Cause of the Problem: Official Investigation
http://www.post-gazette.com/stories/local/breaking/upmc-sued-over-botched-kidney-transplant-315580/ http://www.nbcnews.com/id/44599555/#.UmMHMWRDszQ
• Jan 26: Ms. Mecannic gets her blood work positive for Hepatitis C virus.
• March 29: Second attempt to test for Hepatitis C virus in Ms. Mecannic.
• Several meetings of the transplant team -- they fail to notice the problem. (alert fatigue?)
• April 6: Transplant day!• May 6: Couple learned about botched
transplant.
30
"Between 2007 and 2010, the CDC conducted 200 investigations into potential transmission of HIV and hepatitis B and C due to organ transplants.”
Can we Prevent such life threatening incidents?
http://www.nbcnews.com/id/44599555/#.UmMHMWRDszQ
Over 28,000 organs of all types are transplanted every year in United States alone
31
How could Smart Data help?
Value: Healthcare Provider Context
32
Clinical Decision Making is Complex!
“Health professionals are required to make decisions with multiple foci (e.g. diagnosis, intervention, interaction and evaluation), in dynamic contexts, using a
diverse knowledge base (including an increasing body of evidence-based literature), with multiple variables and individuals involved.”
http://researchoutput.csu.edu.au/R/?func=dbin-jump-full&object_id=9063&local_base=GEN01-CSU01
33
Stakes are high for both doctors and patients!
http://researchoutput.csu.edu.au/R/?func=dbin-jump-full&object_id=9063&local_base=GEN01-CSU01
34
Multimodal, Multisensory, and Multi-organizational Observations
Population health record
Personal health recordExpert opinion Clinical research
What is the overall health of the person?What are the vulnerabilities for organ transplant?
Clinical decision support
http://www.rugeleypower.com/electricity-generation/producing-electricity.php
35
Patient Health Score (diagnostic)
Semantic Perception and risk assessment algorithms can transform raw data (hard to comprehend) to abstractions (e.g., Patient Health is 3 on a scale of 5) that
is intuitively understandable and valuable for decision makers.
Having health score for various patients will allow efficient utilization of a decision maker’s precious attention
Risk assessment model
Semantic Perception
Population health record
Personal health record
Expert opinion
Clinical research
Clinical decision support
36
Patient Vulnerability Score (prognostic)
The Clinical Decision Support systems such as EMR alert system in its current state follows the high recall philosophy by reporting every
possible alert!
Doctors need actionable information and not a deluge of alerts to make timely and important decisions. Providing a vulnerability score would
facilitate right use of Doctor’s time to investigate further on vulnerabilities.
Risk assessment model
Semantic Perception
Population health record
Personal health record
Expert opinion
Clinical research
Clinical decision support
37
Value: Patient Context
How could Smart Data help?
38
3.4 billion people will have smartphones or tablets by 2017 -- Research2Guidance
“Intelligence at the Edges” of Digital Health
http://www.digikey.com/us/en/techzone/energy-harvesting/resources/articles/zigbees-smart-energy-20-profile.html
m-health app market is predicted to reach $26 billion in 2017 -- Research2Guidance
39
Data Overload for Patients/health aficionados
Providing actionable information in a timely manner is crucial to avoid information overload or fatigue
Sleep dataCommunity data
Personal Schedule Activity data
Personal health records
40
Optimizing Cost, Benefit, and Preferences
Algorithms on the patient side should consider all the health signals and provide actionable and timely information for informed decision making
What are the reasons for my increasing weight?What should I consider before I get a kidney transplant?
Semantic Perception
Personalized optimization
Personalized recommendation
Img: http://marloncarvallovillae.blogspot.com/2011_02_01_archive.html http://www.1800timeclocks.com/icon-time-systems/icon-time-upgrades/icon-time-advanced-pack-upgrade-sb100-pro/
Sleep data
Community data
Personal Schedule
Activity data
Personal health records
41
Annotation of sensor data
SemanticSensor
Web
SemanticPerception
Intelligence
at the Edge
Interpretation of sensor data
Efficient execution onresource-constrained devices1 2 3
3 Primary Issues to be addressed
42
RDF OWL
How are machines supposed to integrate and interpret sensor data?
Semantic Sensor Networks (SSN)
43
W3C Semantic Sensor Network Ontology
Lefort, L., Henson, C., Taylor, K., Barnaghi, P., Compton, M., Corcho, O., Garcia-Castro, R., Graybeal, J., Herzog, A., Janowicz, K., Neuhaus, H., Nikolov, A., and Page, K.: Semantic Sensor Network XG Final Report, W3C Incubator Group Report (2011).
44
W3C Semantic Sensor Network Ontology
Lefort, L., Henson, C., Taylor, K., Barnaghi, P., Compton, M., Corcho, O., Garcia-Castro, R., Graybeal, J., Herzog, A., Janowicz, K., Neuhaus, H., Nikolov, A., and Page, K.: Semantic Sensor Network XG Final Report, W3C Incubator Group Report (2011).
45
W3C Semantic Sensor Network Ontology
Lefort, L., Henson, C., Taylor, K., Barnaghi, P., Compton, M., Corcho, O., Garcia-Castro, R., Graybeal, J., Herzog, A., Janowicz, K., Neuhaus, H., Nikolov, A., and Page, K.: Semantic Sensor Network XG Final Report, W3C Incubator Group Report (2011).
46
Semantic Annotation of SWE
Lefort, L., Henson, C., Taylor, K., Barnaghi, P., Compton, M., Corcho, O., Garcia-Castro, R., Graybeal, J., Herzog, A., Janowicz, K., Neuhaus, H., Nikolov, A., and Page, K.: Semantic Sensor Network XG Final Report, W3C Incubator Group Report (2011).
47
To gain new insight in
patient care &
early indications of
disease
Smart Data in Healthcare
49
… and do it efficiently and at scale
What if we could automate this sense making ability?
50
Making sense of sensor data with
51
People are good at making sense of sensory input
What can we learn from cognitive models of perception?• The key ingredient is prior knowledge
52* based on Neisser’s cognitive model of perception
ObserveProperty
PerceiveFeature
Explanation
Discrimination
1
2
Perception Cycle*
Translating low-level signals into high-level knowledge
Focusing attention on those aspects of the environment that provide useful information
Prior Knowledge
53
To enable machine perception,
Semantic Web technology is used to integrate sensor data with prior knowledge on the Web
54
Prior knowledge on the Web
W3C Semantic Sensor Network (SSN) Ontology Bi-partite Graph
55
Prior knowledge on the Web
W3C Semantic Sensor Network (SSN) Ontology Bi-partite Graph
56
ObserveProperty
PerceiveFeature
Explanation1
Translating low-level signals into high-level knowledge
Explanation
Explanation is the act of choosing the objects or events that best account for a set of observations; often referred to as hypothesis building
57
Explanation
Inference to the best explanation• In general, explanation is an abductive problem; and
hard to compute
Finding the sweet spot between abduction and OWL• Simulation of Parsimonious Covering Theory in OWL-
DL (using the single-feature assumption*)
* An explanation must be a single feature which accounts forall observed properties
Explanation is the act of choosing the objects or events that best account for a set of observations; often referred to as hypothesis building
58
Explanation
Explanatory Feature: a feature that explains the set of observed properties
ExplanatoryFeature ≡ ssn:isPropertyOf∃ —.{p1} … ssn:isPropertyOf⊓ ⊓ ∃ —.{pn}
elevated blood pressure
clammy skin
palpitations
Hypertension
Hyperthyroidism
Pulmonary Edema
Observed Property Explanatory Feature
59
Discrimination is the act of finding those properties that, if observed, would help distinguish between multiple explanatory features
ObserveProperty
PerceiveFeature
Explanation
Discrimination2
Focusing attention on those aspects of the environment that provide useful information
Discrimination
60
Discrimination
Universe of observable properties
To determine which possible observations are most informative, find those observable properties that can discriminate between the set of hypotheses.
ExpectedProperties
Not-applicableProperties
Discriminating
Properties
61
Discrimination
Expected Property: would be explained by every explanatory feature
ExpectedProperty ≡ ssn:isPropertyOf.{f∃ 1} … ssn:isPropertyOf.{f⊓ ⊓ ∃ n}
elevated blood pressure
clammy skin
palpitations
Hypertension
Hyperthyroidism
Pulmonary Edema
Expected Property Explanatory Feature
62
Discrimination
Not Applicable Property: would not be explained by any explanatory feature
NotApplicableProperty ≡ ¬ ssn:isPropertyOf.{f∃ 1} … ¬ ssn:isPropertyOf.{f⊓ ⊓ ∃ n}
elevated blood pressure
clammy skin
palpitations
Hypertension
Hyperthyroidism
Pulmonary Edema
Not Applicable Property Explanatory Feature
63
Discrimination
Discriminating Property: is neither expected nor not-applicable
DiscriminatingProperty ≡ ¬ExpectedProperty ¬NotApplicableProperty⊓
elevated blood pressure
clammy skin
palpitations
Hypertension
Hyperthyroidism
Pulmonary Edema
Discriminating Property Explanatory Feature
64
Resource savings of abstracting sensor data
Orders of magnitude resource savings for generating and storing relevant abstractions vs. raw observations.
Relevant abstractions
Raw observations
65
The Decisions are as Good as the Underlying Coded Knowledge
• How do we know whether we have all possible relationships?
• How do we know which relationships are missing?• How can we efficiently fill the missing relationships?
66
Sujan Perera, Cory Henson, Krishnaprasad Thirunarayan, Amit Sheth, Suhas Nair, 'Semantics Driven Approach for Knowledge Acquisition from EMRs', Special Issue on Data Mining in Bioinformatics, Biomedicine and Healthcare Informatics, Journal of Biomedical and Health Informatics (To Appear)
Knowledge is built by abstracting real world facts, once built it should be able to explain the real world
Semantics Driven Approach for Knowledge Acquisition from EMRs
Explanation Module
Explained?
Yes
NoHypothesis
FilteringHypothesis Generation
Hypothesis with High
Confidence
D
D D
DD
D
Patient Notes
Semantics Driven Approach for Knowledge Acquisition from EMRs
UMLS
68
1. Annotate the EMR documents with given knowledgebase2. Find unexplained symptoms3. Generate hypothesis for unexplained symptoms
1. All disorders in document becomes candidates4. Filter out candidate disorder with high confidence
1. Get disorders which has relationship with unexplained symptom in given knowledgebase
2. Collect the “neighborhood” of the disorders3. Get the intersection of “neighborhood” and candidate
disorders
The Algorithm
D1
D5
D2
D3
D4
S1
D8
D12
D6
D9D10
D2D7
D11
D13
D5Candidate Disease
Is symptom of
rdfs:subClassOf
Candidate Filtering Step
Intuition: “similar disorders manifest similar symptoms”
70
Evaluation
Precision = number of suggested correct relationshipsTotal number of suggested
= 73.09%
Recall = correct relationships found all correct relationships – known correct relationships
= 66.67%
If we do not perform the semantic filtering step, the precision would be 30%. High precision is important since it is hard to find domain experts to validate the generated hypothesis.
71
Through physical monitoring and analysis, our cellphones could act as an early warning system to detect serious health conditions, and provide actionable information
canary in a coal mine
kHealth
knowledge-enabled healthcare
72
kHealth to Manage ADHF (Acute Decompensated Heart Failure)
73
Qualities-High BP-Increased Weight
Entities-Hypertension-Hypothyroidism
kHealth
Machine Sensors
Personal Input
EMR/PHR
Comorbidity risk score e.g., Charlson Index
Longitudinal studies of cardiovascular risks
- Find correlations- Validation - domain knowledge - domain expert
Parameterize the model
Risk Assessment Model
Current Observations-Physical-Physiological-History
Risk Score(Actionable Information)
Model CreationValidate correlations
Historical observations of each patient
Risk Score: from Data to Abstraction and Actionable Information
77
1http://www.nhlbi.nih.gov/health/health-topics/topics/asthma/2http://www.lung.org/lung-disease/asthma/resources/facts-and-figures/asthma-in-adults.html 3Akinbami et al. (2009). Status of childhood asthma in the United States, 1980–2007. Pediatrics,123(Supplement 3), S131-S145.
25 million
300 million
$50 billion
155,000
593,000
People in the U.S. are diagnosed with asthma (7 million are children)1.
People suffering from asthma worldwide2.
Spent on asthma alone in a year2
Hospital admissions in 20063
Emergency department visits in 20063
Asthma
78
Asthma is a multifactorial disease with health signals spanning personal, public health, and population levels.
Real-time health signals from personal level (e.g., Wheezometer, NO in breath, accelerometer, microphone), public health (e.g., CDC, Hospital EMR), and population level (e.g., pollen level, CO2) arriving continuously in fine grained samples potentially with missing information and uneven sampling frequencies.
Variety Volume
VeracityVelocity
Value
Can we detect the asthma severity level?Can we characterize asthma control level?What risk factors influence asthma control?What is the contribution of each risk factor?
sem
antic
s Understanding relationships betweenhealth signals and asthma attacksfor providing actionable information
WHY Big Data to Smart Data: Healthcare example
79
Population Level
Personal
Public Health
Variety: Health signals span heterogeneous sourcesVolume: Health signals are fine grainedVelocity: Real-time change in situationsVeracity: Reliability of health signals may be compromised
Value: Can I reduce my asthma attacks at night?
Decision support to doctorsby providing them with
deeper insights into patientasthma care
Asthma: Demonstration of Value
80
Sensordrone – for monitoring environmental air quality
Wheezometer – for monitoringwheezing sounds
Can I reduce my asthma attacks at night?
What are the triggers?
What is the wheezing level?
What is the propensity toward asthma?
What is the exposure level over a day?
What is the air quality indoors?
Commute to Work
Personal
Public Health
Population Level
Closing the window at homein the morning and taking analternate route to office may
lead to reduced asthma attacks
Actionable Information
Asthma: Actionable Information for Asthma Patients
81
Personal, Public Health, and Population Level Signals for Monitoring Asthma
ICS= inhaled corticosteroid, LABA = inhaled long-acting beta2-agonist, SABA= inhaled short-acting beta2-agonist ; *consider referral to specialist
Asthma Control and Actionable Information
Sensors and their observations for understanding asthma
82
Personal Level Signals
Societal Level Signals
(Personal Level Signals)
(Personalized Societal Level Signal)
(Societal Level Signals)Societal Level Signals
Relevant to the Personal Level
Personal Level Sensors
(kHealth**) (EventShop*)
Qualify QuantifyAction
Recommendation
What are the features influencing my asthma?What is the contribution of each of these features?
How controlled is my asthma? (risk score)What will be my action plan to manage asthma?
Storage
Societal Level Sensors
Asthma Early Warning Model (AEWM)
Query AEWM
Verify & augmentdomain knowledge
Recommended Action
Action Justification
Asthma Early Warning Model
*http://www.slideshare.net/jain49/eventshop-120721, ** http://www.youtube.com/watch?v=btnRi64hJp4
83
Population Level
Personal
Wheeze – YesDo you have tightness of chest? –Yes
Observations Physical-Cyber-Social System Health Signal Extraction Health Signal Understanding
<Wheezing=Yes, time, location>
<ChectTightness=Yes, time, location>
<PollenLevel=Medium, time, location>
<Pollution=Yes, time, location>
<Activity=High, time, location>
Wheezing
ChectTightness
PollenLevel
Pollution
Activity
Wheezing
ChectTightness
PollenLevel
Pollution
Activity
RiskCategory
<PollenLevel, ChectTightness, Pollution,Activity, Wheezing, RiskCategory><2, 1, 1,3, 1, RiskCategory><2, 1, 1,3, 1, RiskCategory><2, 1, 1,3, 1, RiskCategory><2, 1, 1,3, 1, RiskCategory>
.
.
.
Expert Knowledge
Background Knowledge
tweet reporting pollution level and asthma attacks
Acceleration readings fromon-phone sensors
Sensor and personal observations
Signals from personal, personal spaces, and community spaces
Risk Category assigned by doctors
Qualify
Quantify
Enrich
Outdoor pollen and pollution
Public Health
Health Signal Extraction to Understanding
Well Controlled - continueNot Well Controlled – contact nursePoor Controlled – contact doctor
84
Personal Health Score and Vulnerability Score
At Discharge
Health Score Non-compliance Poor economic status
No living assistance
Vulnerability Score
Well Controlled Low
Well Controlled Very low
Not Well Controlled
High
Not Well Controlled
Medium
Poor Controlled Very High
Poor Controlled High
Estimation of readmission vulnerability based on the personal health score
85
Health Signal Extraction Challenges
Social streams has been used to extract many near real-time events
Twitter provides access to rich signals but is noisy, informal, uncontrolled capitalization, redundant,
and lacks context
We formalize the event extraction from tweets as a sequence labeling problem
How do we know the event phrases and who creates the training set? (manual creation is ruled out)
Now you know why you’re miserable! Very High Alert for B-ALLERGEN Ragweed I-ALLERGEN pollen. B-FACILITY Oklahoma I-FACILITY Allergy I-FACILITY Clinic says it’s an extreme exposure situation
Idea: Background knowledge used to create the training set e.g., typing information becomes the label for a concept
86
Health Signal Understanding Challenges
Formalized as a problem of structure extraction of a Bayesian Network
Find the structure that maximize the scoring function Where k indexes over all
possible graph structures
Ehsan Nazerfard, Bayesian Networks: Structure Learning, Topics in Machine Learning, 2011.
Where n is the number of nodesin the network
Huge exponential search space with n
Different structures may result in the same structure score (I-Map)
We use declarative knowledge to choose
between Gi and Gj , and to guide the search
Where Xi represents eachobservation
87
How do we implement machine perception efficiently on aresource-constrained device?
Use of OWL reasoner is resource intensive (especially on resource-constrained devices), in terms of both memory and time
• Runs out of resources with prior knowledge >> 15 nodes• Asymptotic complexity: O(n3)
88
intelligence at the edge
Approach 1: Send all sensor observations to the cloud for processing
Approach 2: downscale semantic processing so that each device is capable of machine perception
Henson et al. 'An Efficient Bit Vector Approach to Semantics-based Machine Perception in Resource-Constrained Devices, ISWC 2012.
89
Efficient execution of machine perception
Use bit vector encodings and their operations to encode prior knowledge and execute semantic reasoning
0101100011010011110010101100011011011010110001101001111001010110001101011000110100111
90
O(n3) < x < O(n4) O(n)
Efficiency Improvement
• Problem size increased from 10’s to 1000’s of nodes• Time reduced from minutes to milliseconds• Complexity growth reduced from polynomial to
linear
Evaluation on a mobile device
91
2 Prior knowledge is the key to perceptionUsing SW technologies, machine perception can be formalized and integrated with prior knowledge on the Web
3 Intelligence at the edgeBy downscaling semantic inference, machine perception can
execute efficiently on resource-constrained devices
Semantic Perception for smarter analytics: 3 ideas to takeaway
1 Translate low-level data to high-level knowledgeMachine perception can be used to convert low-level sensory signals into high-level knowledge useful for decision making
92
D. Cameron, G. A. Smith, R. Daniulaityte, A. P. Sheth, D. Dave, L. Chen, G. Anand, R. Carlson, K. Z. Watkins, R. Falck. PREDOSE: A Semantic Web Platform for Drug Abuse Epidemiology using Social Media. Journal of Biomedical Informatics. July 2013 (in press)
Kno.e.sis - Ohio Center of Excellence in Knowledge-enabled ComputingCITAR - Center for Interventions Treatment and Addictions Research
http://wiki.knoesis.org/index.php/PREDOSE
PREDOSE: Prescription Drug abuse Online-Surveillance and Epidemiology
Bridging the gap between researcher and policy makers
Early identification of emerging patterns and trends in abuse
In 2008, there were 14,800 prescription painkiller deaths*
*http://www.cdc.gov/homeandrecreationalsafety/rxbrief/
PREDOSE: Prescription Drug abuse Online-Surveillance and Epidemiology
• Drug Overdose Problem in US• 100 people die everyday from drug overdoses• 36,000 drug overdose deaths in 2008• Close to half were due to prescription drugs
Gil KerlikowskeDirector, ONDCP
Launched May 2011
PREDOSE: Bringing Epidemiologists and Computer Scientists together
Early Identification and Detection of Trends
Access hard-to-reach Populations
Large Data Sample Sizes
Group Therapy: http://www.thefix.com/content/treatment-options-prison90683
Interviews
Online Surveys
Automatic Data Collection
Not Scalable
Manual Effort
Sample Biases
Epidemiologist
Qualitative Coding
Problems
Computer Scientist
Automate Information Extraction & Content Analysis
I was sent home with 5 x 2 mg Suboxones. I also got a bunch of phenobarbital (I took all 180 mg and it didn't do shit except make me a walking zombie for 2 days). I waited 24 hours after my last 2 mg dose of Suboxone and tried injecting 4 mg of the bupe. It gave me a bad headache, for hours, and I almost vomited. I could feel the bupe working but overall the experience sucked.
Of course, junkie that I am, I decided to repeat the experiment. Today, after waiting 48 hours after my last bunk 4 mg injection, I injected 2 mg. There wasn't really any rush to speak of, but after 5 minutes I started to feel pretty damn good. So I injected another 1 mg. That was about half an hour ago. I feel great now.
Codes Triples (subject-predicate-object)
Suboxone used by injection, negative experience Suboxone injection-causes-Cephalalgia
Suboxone used by injection, amount Suboxone injection-dosage amount-2mg
Suboxone used by injection, positive experience Suboxone injection-has_side_effect-Euphoria
experience sucked
feel pretty damn good
didn’t do shit
feel great
Sentiment Extraction
bad headache
+ve
-ve
Triples
DOSAGE PRONOUN
INTERVAL Route of Admin.
RELATIONSHIPS SENTIMENTS
DIVERSE DATA TYPES
ENTITIES
I was sent home with 5 x 2 mg Suboxones. I also got a bunch of phenobarbital (I took all 180 mg and it didn't do shit except make me a walking zombie for 2 days). I waited 24 hours after my last 2 mg dose of Suboxone and tried injecting 4 mg of the bupe. It gave me a bad headache, for hours, and I almost vomited. I could feel the bupe working but overall the experience sucked.
Of course, junkie that I am, I decided to repeat the experiment. Today, after waiting 48 hours after my last bunk 4 mg injection, I injected 2 mg. There wasn't really any rush to speak of, but after 5 minutes I started to feel pretty damn good. So I injected another 1 mg. That was about half an hour ago. I feel great now.
I was sent home with 5 x 2 mg Suboxones. I also got a bunch of phenobarbital (I took all 180 mg and it didn't do shit except make me a walking zombie for 2 days). I waited 24 hours after my last 2 mg dose of Suboxone and tried injecting 4 mg of the bupe. It gave me a bad headache, for hours, and I almost vomited. I could feel the bupe working but overall the experience sucked.
Of course, junkie that I am, I decided to repeat the experiment. Today, after waiting 48 hours after my last bunk 4 mg injection, I injected 2 mg. There wasn't really any rush to speak of, but after 5 minutes I started to feel pretty damn good. So I injected another 1 mg. That was about half an hour ago. I feel great now.
Buprenorphine
subClassOf
bupe
Entity Identification
has_slang_term
SuboxoneSubutex
subClassOf
bupey
has_slang_term
Drug Abuse Ontology (DAO)83 Classes37 Properties
33:1 Buprenorphine24:1 Loperamide
Ontology Lexicon Lexico-ontology Rule-based Grammar
ENTITIESTRIPLES
EMOTIONINTENSITYPRONOUN
SENTIMENT
DRUG-FORMROUTE OF ADM
SIDEEFFECT
DOSAGEFREQUENCY
INTERVAL
Suboxone, Kratom, Herion, Suboxone-CAUSE-Cephalalgia
disgusted, amazed, irritatedmore than, a, few of
I, me, mine, myIm glad, turn out bad, weird
ointment, tablet, pill, filmsmoke, inject, snort, sniffItching, blisters, flushing, shaking hands, difficulty
breathing
DOSAGE: <AMT><UNIT> (e.g. 5mg, 2-3 tabs)
FREQ: <AMT><FREQ_IND><PERIOD> (e.g. 5 times a week)
INTERVAL: <PERIOD_IND><PERIOD> (e.g. several years)
PREDOSE: Smarter Data through Shared Context and Data Integration
Data Type Semantic Web Technique Limitations of Other Approaches
Entity Ontology-driven Identification & Normalization
ML/NLP IR
Requires Labeled Data
Unpredictable term frequencies
Triple Schema-drivenDifficult to
develop language model
Requires entity disambiguation
Sentiment Ontology-assisted Target Entity Resolution
Inconsistent data for Parse Trees or
rules
Diverse simple & complex slang
terms & phrases
PREDOSE: Role of Semantic Web & Ontologies
99
Loperamide-Withdrawal Discovery
Loperamide is used to self-medicate to from Opioid Withdrawal symptoms
100
Big Data from Healthcare Smart Data for Healthcare
Red, yellow, and green indicate high, medium, and low risk allowing decision
makers to focus on red & yellow variables
Big Data vs. Smart Data in Digital Health (Healthcare provider)
Population health record
Personal health record
Expert opinion
Clinical research
Clinical decision support
What is the overall health of the person?What are the vulnerabilities for organ transplant?
Ms. Mecannic’s blood test not yet complete
101
Red, yellow, and green indicating high, medium, and
low risk factors
Recommendation algorithms will analyze data deluge with
domain knowledge
Big Data vs. Smart Data in Digital Health (Healthcare consumer)
Big Data from Healthcare Smart Data for Healthcare
What are the reasons for my increasing weight?What should I consider before I get a kidney transplant?
Sleep data
Community data
Personal Schedule
Personal health records
Activity data
101http://www.airtel.in/forme/important-alerts
Ms. Mecannic: Your blood work is
incomplete. Please finish this before organ donation!
102
• Real Time Feature Streams: http://www.youtube.com/watch?v=_ews4w_eCpg
• kHealth: http://www.youtube.com/watch?v=btnRi64hJp4
• PREDOSE: https://www.youtube.com/watch?v=gCFPzMgEPQM
Demos
103
Take Away
• Data processing for personalized healthcare is lot more than a Big Data processing problem
• It is all about the human – not computing, not device: help them make better decisions, give actionable information– Computing for human experience
• Whatever we do in Smart Data, focus on human-in-the-loop (empowering machine computing!):– Of Human, By Human, For Human– But in serving human needs, there is a lot more than what
current big data analytics handle – variety, contextual, personalized, subjective, spanning data and knowledge across P-C-S dimensions
104
Acknowledgements
• Kno.e.sis team• Funds: NSF, NIH, AFRL, Industry…
• Note:• For images and sources, if not on slides, please see slide notes• Some images were taken from the Web Search results and all such images belong
to their respective owners, we are grateful to the owners for usefulness of these images in our context.
105
• OpenSource: http://knoesis.org/opensource• Showcase: http://knoesis.org/showcase • Vision: http://knoesis.org/vision • Publications: http://knoesis.org/library
References and Further Readings
Amit Sheth’s PHD students
Ashutosh Jadhav
Hemant Purohit
Vinh Nguyen
Lu ChenPavan
KapanipathiPramod
Anantharam
Sujan Perera
Alan Smith
Pramod Koneru
Maryam Panahiazar
Sarasi Lalithsena
Cory Henson
Kalpa Gunaratna
Delroy Cameron
Sanjaya Wijeratne
Wenbo Wang
Kno.e.sis in 2012 = ~100 researchers (15 faculty, ~50 PhD students)
107
thank you, and please visit us at
http://knoesis.org/visionhttp://knoesis.org/amit/hcls
Kno.e.sis – Ohio Center of Excellence in Knowledge-enabled ComputingWright State University, Dayton, Ohio, USA
Smart Data