Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
12/12/2016 ISM 2016,
Medicine or Health!
Today: Medicine
• Reactive
• Broad demographic models
• Medicine based cure
• Real time only in ICU
• ...
Most Fundamental Problem:
Connecting People’s Needs to
Resources Effectively, Efficiently,
and Promptly in given Situation...
Hunter Gatherer:
When needed, you went
to Food.
Now food comes to you.
Until 2000, you went
to a source -- Library.
Now Information finds you.
Disruption in Healthcare
When needed, I go to
the source of
healthcare.
Can healthcare come
to me in time?
Important Revolution in Health
In the mid-20th century, the primary causes of death
worldwide shifted from infections to c...
Personal Health
What’s Lifestyle got to do with it?
Needed: Medical Emancipation
Doctor Knows the Best.
Patient Knows the Best.
Smartphone + Wearables :=
Personalized 24/7 Recording Stethoscope
Event Mining
Machine Learning
Individual Model
used in Cybernetic
Health
‘Likely to get severe heart
attack in 10 minutes ...
Quantifying Lifestyle
Physical
Activities
Popular now: Output and state
Food
Starting to happen:
Input
Environmental
Facto...
“Shallow men believe in luck or in circumstance. Strong men
believe in cause and effect.”
― Ralph Waldo Emerson
What is Cyber Space?
Who invented it?
Animals
Machines
Societies
Published first in 1942
Cybernetics: Real Time Feedback Control
Physical
World
And
Information
Systems
Environment and Resources
Information
Perso...
Cybernetics: Personal
Future Health: Personal, Predictive, Precise,
Persuasive, and Preventive
P5 in Action
Quantifying Lifestyle
Measure
Understand
Control
Improve
If you can't measure it, you
can't control it!
OBJECTIVE SELF (FUTURE WORK)
OBJECTIVE SELF
Anecdotal
Diarizing data
Quantified Self
From data streams to situations
Situations/
Conditions
Event Streams
Observation Streams
Low-level
Analysis
Aggregation
an...
Data Streams
From Data Streams to Abstracted Event
Streams
Data Management
Event Streams
23
Data Streams
Event Stream
DailyLife
Activity
ATUS: American Time Use Survey
Daily Life Activities: How we enter them?
Chronicle of Daily
Life Activities
We collect diverse signals.
Personal Event Streams in a Personicle
Daily Activity
Environment
Medical
Emotion
Activity
Food Food Score
Activity Score
...
Running Example
30
Daily Events
WalkingMeeting Break
Arrive
Home
Exercise Asthma Attack
Pollution Events
91 4 6 10 15 19
L...
Definitions
• Time Interval: [∂, ts, te] ∂+ = ts , ∂− = te
• Semi Interval: [∂+/−, t]
• Point Event (pE): e = (ν, [ E, t])...
Definitions (Cont.)
• Event Stream: ES(i) = {e1
(i), e2
(i), ..., en
(i)}
• Multi-Event Stream: ES = {ES(1), ES(2), ..., E...
Representations
33
Daily Events
WalkingMeeting Break
Arrive
Home
Exercise Asthma Attack
Pollution Events
91 4 6 10 15 19
1...
Pattern Mining Operators
Sequential Co-occurrence
Concurrent Co-occurrence
34
0.25 0.31 0.47 0.11 0.68 0
0.01 0.33 0.64 0.19 0 0.16
0.52 0.12 0.09 0 0.11
0.67 0.1 0 0.7 0.03 0.52
0.13 0 0.75 0.71 0.23...
Cause - Effect Pattern Structure
event1 event2
event3
event4
Effect
Time lag between
events
Sequence of events
Events in p...
Pattern
Mining
High level
Pattern
Formulation
Pattern
Query
…
Data Streams Event Streams Semi-interval Event Sequences
Dat...
User Interface for Interactive Knowledge
Discovery and Model Building
Data-Driven
Hypothesis-
Driven
Pollution and Meteorological Data
Asthma Risk Factor Recognition
1) Pollution increases
suddenly followed by high
wind while temperature
increases slightly ...
Results
Temperature fluctuation has the most impact in the fall and winter seasons
and it is not a risk factor during spri...
Results (Cont.)
• When PM2.5 increases followed by temperature stay high within 3 days,
then asthma outbreak is probable.
...
Spicy Indian food and 2 glasses of wine
result in severe acidity and sleepless nights.
Personicle
Food Stream
t1 t2 t3 t4 ...
Future Health:
• Personalized
• Predictive
• Precise
• Persuasive
• Preventive
Right Moment,
Right Place,
Right Decision,
...
Health Butler
Your food and
activity scores for
the last 3 days are
on low side.
Coming Soon Near You
Current Status
• Institute for Future Health
• Event Framework getting ready
• First Application is Diabetes
– Next Cardio...
ISM 2016 keynote on computing lifestyle 161211
ISM 2016 keynote on computing lifestyle 161211
ISM 2016 keynote on computing lifestyle 161211
Upcoming SlideShare
Loading in …5
×

ISM 2016 keynote on computing lifestyle 161211

1,217 views

Published on

Lifestyle is a major factor in a person's health. By making lifestyle computable, it may be possible to guide health.

Published in: Healthcare

ISM 2016 keynote on computing lifestyle 161211

  1. 1. 12/12/2016 ISM 2016,
  2. 2. Medicine or Health! Today: Medicine • Reactive • Broad demographic models • Medicine based cure • Real time only in ICU • Anecdotal patient data • Doctor knows the best Tomorrow: Health: Health • Predictive and Preventive • personalized and Precise • Lifestyle + medicine • Real time: Womb to Tomb • 24/7 data collection • Patient knows the best
  3. 3. Most Fundamental Problem: Connecting People’s Needs to Resources Effectively, Efficiently, and Promptly in given Situations. 3
  4. 4. Hunter Gatherer: When needed, you went to Food. Now food comes to you.
  5. 5. Until 2000, you went to a source -- Library. Now Information finds you.
  6. 6. Disruption in Healthcare When needed, I go to the source of healthcare. Can healthcare come to me in time?
  7. 7. Important Revolution in Health In the mid-20th century, the primary causes of death worldwide shifted from infections to chronic conditions.
  8. 8. Personal Health What’s Lifestyle got to do with it?
  9. 9. Needed: Medical Emancipation Doctor Knows the Best. Patient Knows the Best.
  10. 10. Smartphone + Wearables := Personalized 24/7 Recording Stethoscope
  11. 11. Event Mining Machine Learning Individual Model used in Cybernetic Health ‘Likely to get severe heart attack in 10 minutes – get him help immediately.’ • Individual Model from data. • Health, social, personal. • Actionable Predictive use. • Better disease models. Healthcare 2020 Objective Self
  12. 12. Quantifying Lifestyle Physical Activities Popular now: Output and state Food Starting to happen: Input Environmental Factors Coming soon: Surround
  13. 13. “Shallow men believe in luck or in circumstance. Strong men believe in cause and effect.” ― Ralph Waldo Emerson
  14. 14. What is Cyber Space? Who invented it? Animals Machines Societies Published first in 1942
  15. 15. Cybernetics: Real Time Feedback Control Physical World And Information Systems Environment and Resources Information Personal Situation and Needs Information Matching Action Signals
  16. 16. Cybernetics: Personal
  17. 17. Future Health: Personal, Predictive, Precise, Persuasive, and Preventive
  18. 18. P5 in Action
  19. 19. Quantifying Lifestyle Measure Understand Control Improve If you can't measure it, you can't control it!
  20. 20. OBJECTIVE SELF (FUTURE WORK) OBJECTIVE SELF Anecdotal Diarizing data Quantified Self
  21. 21. From data streams to situations Situations/ Conditions Event Streams Observation Streams Low-level Analysis Aggregation and Classification 22 Understanding Verification
  22. 22. Data Streams From Data Streams to Abstracted Event Streams Data Management Event Streams 23 Data Streams
  23. 23. Event Stream
  24. 24. DailyLife Activity ATUS: American Time Use Survey
  25. 25. Daily Life Activities: How we enter them? Chronicle of Daily Life Activities
  26. 26. We collect diverse signals.
  27. 27. Personal Event Streams in a Personicle Daily Activity Environment Medical Emotion Activity Food Food Score Activity Score Emotion Score Medical Score Environment Score Health Score = F (Food, Activity, Emotion, Medical, Environment)
  28. 28. Running Example 30 Daily Events WalkingMeeting Break Arrive Home Exercise Asthma Attack Pollution Events 91 4 6 10 15 19 Low Temperature Events Increase Suddenly High Decrease SteadilyMedium Low 7 14 Leave Home Heart rate Events NormalHigh Elevated
  29. 29. Definitions • Time Interval: [∂, ts, te] ∂+ = ts , ∂− = te • Semi Interval: [∂+/−, t] • Point Event (pE): e = (ν, [ E, t]) • Interval Event (iE): e = (v, [E, ts, te]) • Semi-interval Event (sE): e = (v, [E+/−, t]) 31 Life Events WalkingMeeting Break Arrive Home Exercise Asthma Attack 91 4 6 10 15 197 14 Leave Home
  30. 30. Definitions (Cont.) • Event Stream: ES(i) = {e1 (i), e2 (i), ..., en (i)} • Multi-Event Stream: ES = {ES(1), ES(2), ..., ES(∣I∣)} • Pattern: ρ = (X1 ⊙1 X2 ⊙2 ... ⊙k−1 Xk ) Xi ∈ { pE, iE, sE } and ⊙i ∈ { ; , ;ω∆t , , ∣ } 32 Life Events WalkingMeeting Break Arrive Home Exercise AsthmaAttack 91 4 6 10 15 197 14 Leave Home ( Meeting ; Break) ( Exercise+ ;ω[5] AsthmaAttack) ( Walking ; ArriveHome)
  31. 31. Representations 33 Daily Events WalkingMeeting Break Arrive Home Exercise Asthma Attack Pollution Events 91 4 6 10 15 19 1 Low 8 1 0 24 Temperature Events Increase Suddenly High 1 Decrease SteadilyMedium Low 247 1 4 7 14 Leave Home (( Exercise+ Pollution.High) ;ω[5] AsthmaAttack) (( Exercise+ Temp.Low ) ;ω[5] AsthmaAttack) (( Exercise+ (Pollution.High | Temp.Low )) ;ω[5] AsthmaAttack)
  32. 32. Pattern Mining Operators Sequential Co-occurrence Concurrent Co-occurrence 34
  33. 33. 0.25 0.31 0.47 0.11 0.68 0 0.01 0.33 0.64 0.19 0 0.16 0.52 0.12 0.09 0 0.11 0.67 0.1 0 0.7 0.03 0.52 0.13 0 0.75 0.71 0.23 0.25 0 0.43 0.66 0.1 0.2 0.35 Δt E1 E2 E3 E4 E5 E6 E1 E2 E3 E4 E5 E6 0.88 Co-occurrence Matrix Visualization 0.25 0.31 0.52 0.11 0.23 00.33 0.23 0.4 0.28 0 0.23 0.1 0.45 0 0.11 0.56 0.12 0 0.45 0.4 0.52 0.23 0 0.12 0.23 0.31 0 0.23 0.56 0.1 0.33 0.25 0.280.82 0.82 E1 E2 E3 E4 E5 E6 E1 E2 E3 E4 E5 E6 35
  34. 34. Cause - Effect Pattern Structure event1 event2 event3 event4 Effect Time lag between events Sequence of events Events in parallel Cause 1 Cause 2 Cause 3 (No medication ; Exercise)  Asthma attack (Exercise Pollen high)  Asthma attack (Exercise (Pollen high | pollution high))  Asthma attack (Exercise Pollution high)  Asthma attack Exercise  Asthma attack Δt Formulate and query complex patterns:
  35. 35. Pattern Mining High level Pattern Formulation Pattern Query … Data Streams Event Streams Semi-interval Event Sequences Data-Driven Analysis Hypothesis-Driven Analysis While air pressure is high, pollution starts increasing gradually, within T time units asthma outbreak happens. ((pollution_inc_steadily ;ωT asthma_outbreak ) || airpressure_stayshigh) Example: Interactive Visualization Interactive Event Mining
  36. 36. User Interface for Interactive Knowledge Discovery and Model Building Data-Driven Hypothesis- Driven
  37. 37. Pollution and Meteorological Data
  38. 38. Asthma Risk Factor Recognition 1) Pollution increases suddenly followed by high wind while temperature increases slightly will cause an asthma Outbreak within 2 days. 2) Thunderstorm followed by temperature decreases steadily will cause an asthma outbreak within 1 day.
  39. 39. Results Temperature fluctuation has the most impact in the fall and winter seasons and it is not a risk factor during spring or summer During spring and summer, when rain suddenly increases to a very high level, an asthma outbreak is more probable The effect of PM2.5 is not noteworthy in the fall and winter seasons.
  40. 40. Results (Cont.) • When PM2.5 increases followed by temperature stay high within 3 days, then asthma outbreak is probable. • When wind decreases followed by PM2.5 increases within 5 days, then asthma outbreak is probable. • When rain increases followed by PM2.5 stay low within 4 days then an asthma outbreak is probable.
  41. 41. Spicy Indian food and 2 glasses of wine result in severe acidity and sleepless nights. Personicle Food Stream t1 t2 t3 t4 t5 Warn him when he is at an Indian Restaurant.
  42. 42. Future Health: • Personalized • Predictive • Precise • Persuasive • Preventive Right Moment, Right Place, Right Decision, Right way.
  43. 43. Health Butler Your food and activity scores for the last 3 days are on low side. Coming Soon Near You
  44. 44. Current Status • Institute for Future Health • Event Framework getting ready • First Application is Diabetes – Next Cardiovascular and Asthma • Several partners: UCSD, Western Sydney, City University Hong Kong, University of Tokyo, EPFL, … • Expect to release first version of Health Butler in 2nd quarter, 2017.

×