'

Toss ‘N’ Turn: Smartphone as Sleep and Sleep Quality Detector

Понравилась презентация – покажи это...





Слайд 0

Toss ‘N’ Turn: Smartphone as Sleep and Sleep Quality Detector Jun-Ki Min (loomlike@cs.cmu.edu) Afsaneh Doryab Jason Wiese Shahriyar Amini John Zimmerman Jason I. Hong


Слайд 1

Sensing Sleep for… Personal informatics UbiComp system Health monitoring 2


Слайд 2

Current Practices 3


Слайд 3

Opportunities We already have smartphones 83% of millennials sleep with their phone 4 Pew Internet


Слайд 4

How well a smartphone can sense sleep without requiring changes in our behavior? 5 Task 1. Detect bedtime, waketime and duration Task 2. Infer daily sleep quality Task 3. Classify good or poor sleeper


Слайд 5

Toss ‘N’ Turn (Data Collection Ver.) 6


Слайд 6

Modeling 7 Sound Motion Sleep


Слайд 7

User Study Recruited good and poor sleepers Living in US, age > 18 Pay $2 USD for each diary entry (a maximum $72) Collected sleep data for a month 30 participants signed up and 27 completed Total 795 sleep-diary entries 8


Слайд 8

Ground Truthing 9 User Study Global score > 5 indicates a subject is having poor sleep Subjective sleep quality + Sleep latency + Sleep efficiency + Sleep duration + Use of medication + Sleep disturbances ---------------------------------- = Global sleep quality


Слайд 9

Demographics 10 User Study 11 Share bed with 3 8 3 1 1 Disrupting noises in the bedroom 12 15 Yes No Age 10 10 5 1 1 20 30 40 50 ? Regularly work 22 5 No Yes Poor sleeper (PSQI global score > 5) Good sleeper (PSQI global score ? 5)


Слайд 10

Evaluation Classifier Bayesian network (BN) with correlation-based feature selection Task 1. Detect bedtime, waketime and duration Task 2. Infer daily sleep quality Train the model individually (leave-one-day-out cross validation) Task 3. Classify good or poor sleeper Leave-one-person-out cross validation 11


Слайд 11

Task 1: Sleep Detection Detect sleep windows ? Detect sleep time 94.5% in classifying sleep/not-sleep windows Evaluation Bedtime detection Baseline (avg. time) Our method Waketime detection Baseline (avg. time) Sleep duration inference Baseline (avg. time) -150 150 -120 120 90 60 30 0 -30 -60 -90 Average minutes of over (+) and under (-) estimation errors Our method Our method


Слайд 12

Task 2: Daily Sleep Quality Inference Evaluation Detect sleep ? Classify the quality of sleep 84.0% in classifying good/poor sleeps Accuracy (%) Our method Random Poor sleep detection (F-score)


Слайд 13

Task 3: Good/Poor Sleeper Classification Evaluation Infer daily qualities ? Classify the sleeper type 81.5% in classifying good/poor sleepers Our method Random Accuracy (%) Poor sleeper detection (F-score)


Слайд 14

Discussion How well a smartphone can sense sleep without requiring changes in our behavior? Task 1. Detect bedtime, waketime and duration within 35, 31, and 49 minutes of errors, respectively Task 2. Infer daily sleep quality with 84% accuracy Task 3. Classify good or poor sleeper with 81% accuracy 15


Слайд 15

Top Five Features Time Battery charging / not-charging Min. movement Std. sound amplitude Q3 sound amplitude Bedtime Waketime Sleep duration Std. movement Yesterday’s sleep quality 16 Discussion Sleep detection Sleep quality inference


Слайд 16

Sleep Detection Errors 17 Discussion People who sleep alone People who have a sleep partner Phone location Error (minutes)


Слайд 17

General vs. Individual Models Sleep detection: 93.06% vs. 94.52% Need 3 days of ground truthing to train an individual model Sleep quality inference: 77.23% vs. 83.97% Need 3 weeks of ground truthing to train an individual model 18 Discussion


Слайд 18

Limitations Subjective vs. objective sleep quality “How was your sleep last night? Rate it on a one to five scale score” does not capture the full extent of a sleep session People tend to over / underestimate their sleep Small sample size of poor-quality sleep 19 Discussion


Слайд 19

Thanks! More info at cmuchimps.org or email loomlike@cs.cmu.edu Special thanks to: DARPA, Google


Слайд 20

Backup Slides 21


Слайд 21

Data Collection Frequency 22


Слайд 22

Features 23


Слайд 23

Modeling: Data Processing Rational for “10 minutes” Level of granularity when participants report sleep time Median sleep latency = 10.9 minutes 90,097 windows, 711 not-sleep and 728 sleep segments 24 … Sound Motion 10-minute window Sleep Sleep Not-sleep Not-sleep Bedtime Waketime


Слайд 24

Sleep Detection & Quality Inference 25 Modeling … Sound Motion 10-minute window Sleep …0011000001010000011111111011010100001010000100000…


Слайд 25

Infer Other Contexts Sleep alone vs. with others 84.2% Phone on the bed vs. near the bed vs. elsewhere 91.9% 26


×

HTML:





Ссылка: