BINGO! DINO DNA! HEALTH IDENTITY FROM THE WRIST BRIAN WILT HEAD OF - - PowerPoint PPT Presentation

▶
bingo dino dna
SMART_READER_LITE
LIVE PREVIEW

BINGO! DINO DNA! HEALTH IDENTITY FROM THE WRIST BRIAN WILT HEAD OF - - PowerPoint PPT Presentation

Dec 12, 2022 330 likes •990 views

BINGO! DINO DNA! HEALTH IDENTITY FROM THE WRIST BRIAN WILT HEAD OF DATA SCIENCE AND ANALYTICS @BRIANWILT NOV 16 2015 THEY TOLD ME TO SLEEP NORMAL BRIDGING SCIENCE AND ENGINEERING UP 3 ROLE OF DATA+ML AT JAWBONE SCIENCE

slide-1
SLIDE 1

BINGO! DINO DNA!

HEALTH IDENTITY FROM THE WRIST

BRIAN WILT HEAD OF DATA SCIENCE AND ANALYTICS @BRIANWILT

slide-2
SLIDE 2

NOV 16 2015

slide-3
SLIDE 3

“THEY TOLD ME TO SLEEP NORMAL”

slide-4
SLIDE 4
slide-5
SLIDE 5

BRIDGING SCIENCE AND ENGINEERING

slide-6
SLIDE 6
slide-7
SLIDE 7

UP3

slide-8
SLIDE 8
  • SCIENCE
  • PERSONALIZATION
  • MOTIVATION

ROLE OF DATA+ML AT JAWBONE

slide-9
SLIDE 9

SCIENCE: UNDERSTANDING THINGS NEVER BEFORE SEEN

slide-10
SLIDE 10

0% 25% 50% 75% 100%

9p 10p 11p 12a 1a 2a 3a 4a 5a 6a 7a 8a 9a

0-15 mi 0-25 mi 25-50 mi 50-75 mi 75-100 mi

NAPA, SONOMA SAN FRANCISCO, OAKLAND NAPA, SONOMA, BERKELEY

SOUTH NAPA EARTHQUAKE

slide-11
SLIDE 11

HEALTH WITH SHARED CONTEXT IN REAL TIME

slide-12
SLIDE 12

PERSONALIZATION: AN EXPERIENCE THAT KNOWS YOU

slide-13
SLIDE 13

YouTube was abuzz with viral videos of small children — yet to speak, read or write — “pinching” magazine articles with their fingers as they would an iPad. These children were heralded as […] “digital natives”. [Now is a] new revolution, this time of “data natives” who expect their world to be “smart” and seamlessly adapt to them and their taste and habits.

RISE OF THE DATA NATIVES

MONICA ROGATI

slide-14
SLIDE 14
slide-15
SLIDE 15

MOTIVATION: A SMART COACH HELPING YOU LIVE HEALTHIER

slide-16
SLIDE 16 HEALTHY WEIGHT

FEMALE MALE

7.2 HRS SLEEP 6.2 7 6.8 6.6 6.4

WEIGHT AFFECTS SLEEP

19 23 27 31 35 39 43 47 (BMI)
slide-17
SLIDE 17

logged 75% more meals per week logged 60% more workouts per week beat their step goal 43% more often averaged 17% more steps per day logged 16% more weigh-ins per week

UP USERS WITH MAJOR WEIGHT LOSS

slide-18
SLIDE 18 IDENTITY BIOMETRICS ACTIVITIES

A COMPLETE PICTURE OF YOU FULLY CONTEXTUALIZED

24 7 DATA

slide-19
SLIDE 19

HOW DO WE MAKE THIS HAPPEN?

slide-20
SLIDE 20

DATA PRODUCT ALL-STARS

slide-21
SLIDE 21

DATA PRODUCT ALL-STARS

NEWS FEED PEOPLE YOU
 MAY KNOW PAGERANK RELATED ITEMS

slide-22
SLIDE 22

DATA PRODUCT ALL-STARS

NEWS FEED PEOPLE YOU
 MAY KNOW PAGERANK RELATED ITEMS

CTR

slide-23
SLIDE 23

CHRONIC DISEASE CARE IS 86% OF US HEALTHCARE COSTS

DIABETES

$176 BN

🍱

CARDIOVASCULAR

$193 BN

💕

CANCER

$157 BN

🚭

OBESITY

$147 BN

🍠

67% OF GYM MEMBERSHIPS GO UNUSED

slide-24
SLIDE 24

HOW WILL WE MAKE HEALTH DATA PRODUCTS?

slide-25
SLIDE 25

DATA SCIENCE FOR THE LITTLE GUY

slide-26
SLIDE 26

DATA SCIENCE FOR THE LITTLE GUY

PROBLEM LARGE COMPANY SMALL COMPANY PEOPLE

Where do I put my data science team? Vertical/horizontal organization? Only one or two people LOL

PRODUCT

Established Nascent

DATA

A mess A mess

ML

Deep on solving a problem, predictive Understanding and interpreting

slide-27
SLIDE 27

DATA SCIENCE FOR HARDWARE

SIGNALS

Raw + Rich Compressed Limited

CONTEXT

Limited Sensor fusion Historical + Population

USERS

Single Single Aggregate

LATENCY

Seconds Minutes Slow

COMPUTE

Limited Powerful HAL 9000

DEPLOYMENT

Months Weeks Hours

slide-28
SLIDE 28 Awesome! Looks like you racked up some steps. What were you up to? BASKETBALL TENNIS ZUMBA +5,972 steps | +50%of goal

54m activity

Edit

ACTIVITY CLASSIFICATION

VERSION 0 MOST COMMON WORKOUT

58% ACCURACY

VERSION 2 SERVER-SIDE MODEL

85% ACCURACY

VERSION 1 LAST WORKOUT

65% ACCURACY

VERSION 3 BAND-BASED MODEL

99% ACCURACY

slide-29
SLIDE 29

THE FIRMWARE HAS TO WORK THE FIRST TIME

slide-30
SLIDE 30

TESLA HAS TO DELIVER THE CAR FIRST

slide-31
SLIDE 31

DATA SCIENCE FOR WEARABLES

slide-32
SLIDE 32
slide-33
SLIDE 33

WE CAN GET EXPONENTIALLY MORE FROM THE WRIST

UP

2011

Accelerometer

UP3

2015

Accelerometer Temperature GSR Heart Rate

slide-34
SLIDE 34

WE CAN GET EXPONENTIALLY MORE FROM THE WRIST

UP

2011

Accelerometer

UP3

2015

Accelerometer Temperature GSR Heart Rate

GERMAN PEDOMETER 1590

slide-35
SLIDE 35

THE IPHONE MOMENT IS COMING

slide-36
SLIDE 36

SLEEP 330M STEPS 4.1T MEALS 270M WORKOUTS 100M

slide-37
SLIDE 37

HEALTH DATA IS WIDE

slide-38
SLIDE 38

HEALTH DATA IS WIDE

slide-39
SLIDE 39

BUT WIDE DATA TAKES MORE CLEANING

  • CLEAN EARLY AND OFTEN
  • APIS ARE HARD
  • BETTER TOOLS ARE NEEDED
  • ML IS NAIVE
slide-40
SLIDE 40 50 60 70 80 11:00 AM 11:10 AM 11:20 AM 11:30 AM 11:40 AM 11:50 AM 12:00 PM 12:10 PM 12:20 PM 12:30 PM

CONTEXT IS KING

EMPLOYEE HEART RATE DURING LAST JAWBONE ALL HANDS

slide-41
SLIDE 41

SCALE GRACEFULLY

slide-42
SLIDE 42

BEDTIME

slide-43
SLIDE 43

BEDTIME

slide-44
SLIDE 44

16 MIN

slide-45
SLIDE 45

SLEEP RECOVERY

slide-46
SLIDE 46

BUT WHAT ABOUT DATA WE CAN’T GET

slide-47
SLIDE 47
  • API INTEGRATIONS
  • BAYESIAN INFERENCE
  • PUBLIC WHO/CENSUS DATA
  • USER DATA
  • UNSTRUCTURED TEXT
  • CREATIVITY
  • HIRING
slide-48
SLIDE 48

DATA PRODUCTS SCALE HUMAN JUDGMENT.

THEY ALSO SCALE AND REINFORCE HUMAN BIAS. @CLARECORTHELL

slide-49
SLIDE 49

Tokyo office workers slept at least 30 minutes less than their counterparts in New York, Paris, Shanghai and Stockholm each night, averaging six hours, or 14 percent less than the recommended minimum, a study said. The study surveyed 180 men and women from each of the five cities.

TOKYO’S WORKERS GET LESS SLEEP THAN US, ASIAN COUNTERPARTS

KANOKO MATSUYAMA (2011)

slide-50
SLIDE 50

datazoo=# SELECT gender, COUNT(*), AVG(total_hrs) datazoo-# FROM up_mod.sleep_clean_night datazoo-# WHERE gender IS NOT NULL GROUP BY gender; gender | count | avg

  • -------+-----------+---------

M | 147284397 | 6.77170 F | 165579431 | 7.11274 Time: 1333.536 ms

DATA DEMOCRATIZATION IS TABLE STAKES

slide-51
SLIDE 51

STREAM API EVENTING USER DATA

DYNAMO

REDSHIFT ANALYTICS

DATA PRODUCTS PIPE

slide-52
SLIDE 52

EXPERIENCE FIRST, THEN TECH

slide-53
SLIDE 53

NOT JUST ANALYTICS — DEMOCRATIZE DATA PRODUCTS

slide-54
SLIDE 54

NOT TRACKING, BUT COACHING

slide-55
SLIDE 55 Not today Early to Bed Smart Coach noticed that you’ve been turning in late recently. Remember, your brain needs plenty of pillowtime to sort new information. Get in bed by 10:50pm tonight? SMART COACH I’m in
slide-56
SLIDE 56 Not today Early to Bed Smart Coach noticed that you’ve been turning in late recently. Remember, your brain needs plenty of pillowtime to sort new information. Get in bed by 10:50pm tonight? SMART COACH I’m in Not today Early to Bed Smart Coach noticed that you’ve been turning in late recently. Remember, your brain needs plenty of pillowtime to sort new information. Get in bed by 10:50pm tonight? SMART COACH I’m in
slide-57
SLIDE 57 UP UP slide to view now Today’s move summary: 10,518 steps, covering 5.1 mi, burning 1,518 cal. That’s 105% of your goal.

SMART COACH GUIDES YOUR SLEEP

Not today Early to Bed Smart Coach noticed that you’ve been turning in late recently. Remember, your brain needs plenty of pillowtime to sort new information. Get in bed by 10:50pm tonight? SMART COACH I’m in
slide-58
SLIDE 58 UP UP slide to view now Today’s move summary: 10,518 steps, covering 5.1 mi, burning 1,518 cal. That’s 105% of your goal.

5 DAY

OCT 23 STREAK Averaging 7h 6m per night Not today Early to Bed Smart Coach noticed that you’ve been turning in late recently. Remember, your brain needs plenty of pillowtime to sort new information. Get in bed by 10:50pm tonight? SMART COACH I’m in

SMART COACH GUIDES YOUR SLEEP

slide-59
SLIDE 59 UP UP slide to view now Today’s move summary: 10,518 steps, covering 5.1 mi, burning 1,518 cal. That’s 105% of your goal.

5 DAY

OCT 23 STREAK Averaging 7h 6m per night e been , your
  • sort
y 10:50pm I’m in

SMART COACH GUIDES YOUR SLEEP

slide-60
SLIDE 60 Not today

Early to Bed

Smart Coach noticed that you’ve been turning in late recently. Remember, your brain needs plenty of pillowtime to sort new information. Get in bed by 10:50pm tonight? SMART COACH I’m in

23M

MORE SLEEP, COMPARED TO THE CONTROL GROUP

72%

INCREASED LIKELIHOOD OF BEATING THEIR SLEEP GOAL

UP HELPS USERS SLEEP MORE

slide-61
SLIDE 61

ANDRE IGOUDALA

FINALS MVP

  • CONSISTENT BEDTIME
  • PUSH FOR LONGER SLEEP DURATION
  • TOOLS TO HELP FALL ASLEEP
  • STRATEGIC CAFFEINE USAGE
  • NAPPING RECOMMENDATIONS

SLEEP COACHING

  • HOME: SLEEPS 5H 48M
  • AWAY: SLEEPS 6H 15M
  • DIFFICULTIES
  • FALLING ASLEEP AFTER GAME
  • STAYING ASLEEP
  • NAPS SOMETIMES
  • DRINKS CAFFEINE BEFORE AND

DURING GAMES

SLEEP PROFILE

slide-62
SLIDE 62

+29%

POINTS PER MINUTE

SLEEP HELPED ANDRE’S OFFENSE

FREE THROW

+8.9%

SLEEP IMPROVED ANDRE’S 3P% BY 2.18X

>8 HRS <7 HRS

3 POINT PERCENTAGE 0.545

7-8 HRS

0.462 0.250

SLEEP HELPED ANDRE’S GAME

  • 37%

TURNOVERS PER FOULS

  • 45%
slide-63
SLIDE 63
  • WE DO BIG THINGS AT SMALL COMPANIES
  • MORE IMPORTANT TO FIGURE OUT WHAT

TO DO THAN DO IT

  • DATA SCIENCE FOR “GOOD”
  • IMPROVE HEALTH WITH BIG DATA TOOLS
  • USER EXPERIENCE COMES FIRST
  • THE FIRST ML MODEL ISN’T A MODEL
  • DEMOCRATIZE DATA PRODUCTS

CONCLUSIONS

slide-64
SLIDE 64

BINGO! DINO DNA!

HEALTH IDENTITY FROM THE WRIST

BRIAN WILT HEAD OF DATA SCIENCE AND ANALYTICS @BRIANWILT