Transcript Sleep Tracking and Wearable Tech

Sleep Tracking and Wearable Tech
Anita Valanju Shelgikar, MD
October 20, 2016
Objectives:
Sleep
tracking
Wearable
tech
Clinical
Applications
Key points:
Sleep
tracking
Expect more of our
patients to use wearables
as part of their daily
routine
Wearable
tech
Validation studies for sleep
tracking: some exist, but
we need more
Clinical
Applications
Incorporation of patientspecific data from
wearables may allow
us to better serve
each individual patient
https://en.wikipedia.org
Sleep monitoring
Polysomnography
Recorded channels
Electroencephalogram
Electrooculogram
Chin electromyogram
Electrocardiogram or
heart rate
Airflow
Respiratory effort
Oxygen saturation
Image from https://www.nhlbi.nih.gov
http://www.newgate.uk.com/
Home sleep apnea testing
4-7 recorded channels
Electrocardiogram or heart rate
Airflow + respiratory movement
OR
At least 2 respiratory movement
channels
Oxygen saturation
Images from http://yoursleep.aasmnet.org
www.ppdsleep.com
Actigraphy
Image from www.sleepdt.com/
CHEST 2011; 139( 6 ): 1514 – 1527
Accelerometer = movement monitor
• Measure acceleration along a given axis
•
•
•
Micro–mechanical springs
Changes in capacitance
Piezo–electric
Accelerometer = movement monitor
• Measure acceleration along a given axis
•
•
•
Micro–mechanical springs
Changes in capacitance
Piezo–electric
• Multiple axis measurements can be
bundled into a single monitor,
allowing multiple planes of movement
to be captured
• Counts are summed over a specified
period of time (epoch) and stored
• Sensor converts movements into
electrical signals (counts) that are
proportional to the muscular force
producing motion
• Many accelerometers have the storage
capacity to assess physical activity
over 21 day periods using a 30 or 60
second epoch
Images from www.qsportstechnology.com/
http://www.parcph.org/accDef.aspx
quantifiedself.com/gsr/
www.firstcoastnews.com/
Patient-driven monitoring
Mobile applications (apps)
Images from www.businessinsider.com
Wearable technologies
www.psfk.com
14 IEEE Pulse September/October 2014
• Cellphone accelerometers are based on
semiconductor accelerometers
•
Microchips with all their components chemically
etched onto the surface of a piece of silicon
•
Measure changes in capacitance between
electrodes
https://blogs.windows.com/buildingapps/2010/09/08/using-the-accelerometer-on-windows-phone-7/
Patient-driven monitoring
Mobile applications (apps)
Images from www.businessinsider.com
Wearable technologies
www.psfk.com
What are wearables?
Digital Insoles
Apple Watch
Sony SmartEyeGlass
Wearable technology
- Electronic technologies or computers incorporated into clothing and
accessories
- Smart sensors use of a web connection, usually using Bluetooth to
connect wirelessly to a smartphone
Smarty Ring
Misfit Swarovski Shine
Implantables
Jawbone UP
Samsung Gear VR
www.wareable.com
MBody Bike & Run
NFC Ring
Artificial Intelligence in Medicine 56 (2012) 137–156
Wearable self-tracking devices and
mobile apps
CHEST 2016;
http://dx.doi.org/10.1016/j.chest.2016.04.016
?
Image from www.theatreatfirst.org
“Quantified Self”
• Coined by Wired magazine in 2007
• ‘A collaboration of users and tool makers
who share an interest in self knowledge
through self-tracking.’”
http://www.healthworkscollective.com
Gartner Hype Cycle
5 key phases of a technology's life cycle
Image from www.gartner.com
Stated another way…
https://www.linkedin.com/pulse/20140814105952-7082046-gartner-hype-cycle-2014
Use of patient-driven monitoring
• 2012 study from the Pew Research Center
• Estimated that 7 out of 10 Americans were
tracking health indicators, either for
themselves or for a loved one
• 21% of the respondents claimed to use some
form of technology to record and store these
data
• Given the continued growth of health tracking
technologies, these numbers are now likely
greater than in 2012
J Sleep Res. (2015) 24, 121–123
Use of patient-driven monitoring
• Uptake of wearable devices is bimodally distributed
• 25 to 34-year-olds use them for fitness enhancement
• 55 to 64-year-olds use them to improve overall health
• Worldwide it is estimated that about 100 million people
track some form of fitness or health data on a regular basis
• Hard to know which of these are related to sleep
J Sleep Res. (2015) 24, 121–123
BMC Medicine (2015) 13:77
Patient-driven monitoring
Concerns
• Some apps collect information from individuals and make pooled data publicly
available for consumption, research, or marketing purposes, without each
individual’s enduring consent or knowledge.
• Legal accountability and regulation of consumer health technologies are
developing but not fully implemented.
• FDA oversight of mobile apps is being defined but will need to continually
evolve as technology advances.
• Many health-related technologies marketed directly to users claim to be
“entertainment devices” and deny any official medical claims in fine print,
though the public is generally unaware of this.
J Clin Sleep Med 2015;11(12):1455–1461.
Patient-driven monitoring
Potential Benefits
• Integrated use of mobile apps will allow patient data to be collected more
frequently and more effectively, which may facilitate improvement in quality
of patient care.
• Patient-specific data from patient self-monitoring can be used to make
individualized diagnoses and/or treatment plans.
• Use of mobile apps and wearable tech may facilitate better chronic disease
management via more real-time communication with patient and reduction of
unnecessary office visits.
• Use of patient-driven monitoring could impact population health management
via increased access to care.
N Engl J Med 2014; 371:372-379
“Tonight, I’m launching a new Precision Medicine
Initiative to bring us closer to curing diseases like
cancer and diabetes — and to give all of us access to
the personalized information we need to keep
ourselves and our families healthier.”
President Barack Obama, State of the Union Address,
January 20, 2015
http://www.nih.gov/precisionmedicine
Tech and precision medicine
• Can provide phenotypic
information to complement
available genotypic information
• Example:
– Recent cardiovascular mobile
health study(MyHeartCounts)
recruited 30, 000 smartphone
users in the first2 weeks after
launch
• Potential applications:
– Disease prevention via
continuous monitoring of activity,
sleep, heart rate, etc.
– Immediate, personalized
feedback to help modify behavior
– Assess effectiveness of an
intervention
•
Individual or population
JAMA June 2, 2015 Volume 313, Number 21
Health tracking &
precision medicine
Tangible benefits so far
• 2013 Data from the Pew Research Center indicate that people with more
serious health problems are more likely to report benefits as a result of
tracking their health
• Meta-analysis of activity monitor-based counseling studies with diabetes
patients
• Activity monitor-based counseling had a beneficial effect on physical activity,
blood glucose, systolic blood pressure, and body mass index
• Use of a wearable activity tracker during inpatient post-surgical showed a
significant relationship between the early recovery step count, hospital length
of stay in a group of elderly cardiac surgery patients
Ann Med. 2013;45:397–412.
J Med Internet Res. 2014;16:192
Ann Thorac Surg. 2013;96:1057–61
Health tracking &
precision medicine
Room for improvement
• Review of popular activity monitors showed that that goal-setting, behavioral
goal review, behavior feedback, self-monitoring, and rewards were generally
included
• Often lacked other important components, such as problem-solving,
behavioral instruction, and commitment strategies
• Need more evidence about effectiveness of health-tracking data in the
management of chronic diseases
• A survey of 2,000 individuals showed that 80% expressed concerns about
privacy of personal data with use of wearable tech
BMC Medicine (2015) 13:77
https://newsroom.accenture.
How can I learn
more about
health tracking,
mobile apps,
and wearables?
https://commonbond.co
http://apps.nhs.uk/
The National Health Service (NHS) in England
Local resources
Patricia Anderson
Emerging Technologies Informationist,
Taubman Health Sciences Library
Marc Stephens
Instructional Designer – MSIS
University of Michigan Medical School
“Tech Savvy Fitness”
Key points:
Sleep
tracking
Expect more of our
patients to use wearables
as part of their daily
routine
Wearable
tech
Validation studies for sleep
tracking: some exist, but
we need more
Clinical
Applications
Incorporation of patientspecific data from
wearables may allow
us to better serve
each individual patient
Screenshot of the graphical
display of a whole-night
recording from a study
subject by the Sleep Time app
on an iPhone.
Comparison of absolute parameters obtained by
polysomnography and provided by the app for subjects
in the study (n = 20).
ap
bp
ap
cp
Hypnogram showing sleep stages
recorded by simultaneous
polysomnography.
J Clin Sleep Med 2015;11(7):709–715
< 0.0001
= 0.008
< 0.0001
= 0.015
• The app’s sleep vs. wake discrimination mirrored that reported for wrist
actigraphy:
• ~90% sensitive and ~50% specific for sleep
• The app overestimated sleep (as does wrist actigraphy)
• Likely because quiet wakefulness contains little movement
Conclusion:
• Sleep Time app is insufficiently precise to serve as an
alternative to polysomnography for sleep- wake
cycle analysis
• May serve as an alternative to actigraphy or
subjective measurements of sleep (e.g. sleep diaries)
J Clin Sleep Med 2015;11(7):695–696
J Clin Sleep Med 2015;11(7):709–715
Image from www.respitek.com.tr/
Validation data – activity monitor
Fitbit vs PSG
Actigraph vs PSG
Fitbit vs Actigraph
24 healthy adults
- Fitbit and Actigraph worn during
polysomnography
Data collection
- Fitbit: 60 second epochs
- Actigraph: 30 second epochs
Conclusion:
Fitbit and Actigraph
- Overestimated sleep efficiency and total
sleep time (overcalling quiet wakefulness)
- High sensitivity for accurately identifying
sleep within all stages of sleep
- Low specificity for accurately identifying
wake
Sleep Breath (2012) 16:913–917
Image from www.fitbit.com
Validation of another wearable
Sixty-five adolescents ages 12–22 years (28 females)
SLEEP 2015;38(9):1461–1468
Image from http://activitytrackerworld.com/jawbone-up4-review/
Validation data for a wireless system
29 healthy adults
- Concurrent
polysomnography,
actigraphy, and wireless
system (Zeo) in a sleep
laboratory for one night
[preceded by an acclimation
night]
J. Sleep Res. (2012) 21, 221–230
*Performance negatively
impacted by bruxism and
by neuroactive
medications
- Altered forehead-based
signal properties and
algorithm output
accuracy
J Clin Sleep Med 2015;11(7):695–696
Neurology 2012;2012:153745
“Yet while it may have been the darling of
sleep trackers, ‘consumers weren’t that
interested in complex and relatively costly
methods for tracking sleep,’ analyst Michael
Gartenberg told Wired.
Zeo sold for about $400 and required users to
wear an uncomfortable elastic headband to
sleep that by morning had left unsightly
indentations on your forehead.”
http://www.wired.com/2013/03/lights-out-for-zeo/