ORCATECH The Internet of Everything – Pervasive Computing for Health

Jeffrey KayeLayton Professor of Neurology & Biomedical EngineeringOregon Center for Aging & TechnologyLayton Aging & Alzheimer's Disease Center kaye@ohsu.edu

Carl Richards

OREGON CONNECTIONS TELECOMMUNICATIONS CONFERENCE - “Broadband: The Pulse of the Future”

October 23-24, 2013, Hood River, Oregon

Current Assessmentis Limited

Cardinal features of health change - slow decline punctuated with acute events - are challenging to assess with current tools and methods.

Assessment LimitationsInclude…

• Rely on sparsely spaced, brief queries – questionnaires, phone or in-person exams.

• Performed at the convenience of the assessor. • Depend on recall of events or brief snap-shots of

function.• Use artificial or non-real world tests; not fun• Assumes observations recorded during the exam

represent typical function• Challenged to track across the course of illness.• High inter-rater or test-to-test variability• Limited knowledge of other events that can significantly

effect outcomes (e.g., sleep, socialization, physical activity)

Early detection

The Greatest Challenge: Detecting Meaningful Change

BaselineBaseline 3 years3 years 6 years6 years

Mea

sure

Symptoms Reported

Functional range

Change

Path Forward: Change the Paradigm

New Discovery New & Transformed Businesses

New Discovery New & Transformed Businesses

OPTIMAL HEALTHOPTIMAL HEALTH

Pervasive Computing Wireless Technologies

“Big Data” Analytics

Pervasive Computing Wireless Technologies

“Big Data” Analytics

• Real-time• Continuous• Home-

based• Unobtrusive• Ambient

• Real-time• Continuous• Home-

based• Unobtrusive• Ambient

To Facilitate this Change…

• Target key functions most related to QoL and highest costs that can be unobtrusively detected to change over time (e.g., cognition, gait, mood, pain, sleep, socialization).

• Build upon advances in remote sensing, pervasive computing, telehealth, activity and behavior modeling creating an ambient, multi-domain home-based assessment system.

• Minimize need to wear, carry or tend to devices, and especially to disrupt the person’s usual daily routine.

• Increase sensitivity by continuous, multi-domain assessment.

Doorsensors

Walking sensors

Activity sensors

Cell phone as prompting device and for location

tracking

Localization sensors

Actigraphy devices

Bed Sensors

Medication tracking device

Phone sensors

Physiological sensors

Pervasive Computing Platform Elements

7

PC/Kiosk/Etc.: Experience sampling; cognitive testing; social engagement; coaching

What you get: Continuous, Holistic Assessment “A Day in the Life”

Embedded Ambient Assessment Technologies

SecureInternet

SecureInternet

Activity Time & Location Gait

Speed

Doors Opening/Closing

Computer/KioskActivity

PhoneActivity

MedTrackerBody

Composition

Together: A Community-wide Home-Based Assessment Platform

Kaye et al. Journals of Gerontology: Psychological Sciences, 2011 10

Identifying change using remote assessment methods: Evidence Examples

Room Transitions during a Norovirus Epidemic

Intact MCI

Daily Activity &

Cognitive Decline

Total Activity: Life Space & EventAnalysis

Spiral plot: The plot is a 24 hour clock representing here 8 weeks of continuous data. At the top of the clock is midnight; at the bottom is noon. Each concentric blue circle outward represents 2 weeks of time. The colors of the dots represent firings of sensors by location

Norovirus Epidemic: All ill patients identified by decreased room transition events without self report

-34%-34%

-24%-24%

-40%-40%

Campbell, 2011

Walks: From 2 to 7000 per year

Photo: NYT, 2009

Hayes, 2009; Hagler, 2010; Kaye, 2012

Variability in walking speed and total activity differentiates MCI from cognitively normal people

MCI

NL

• Mean age = 88 years• Mean in-home motion-activity

sensing 315 ± 82 days

Hayes et al. Alzheimer's & Dementia, 2008; 4(6): 395-405.Hayes et al. Alzheimer’s & Dementia, 2008

Routine home PC use over time (without formal tests or queries) detects change in those with MCI

• Mean 1.5 hours on computer/per day at baseline month

• Over time:– Less use days

per month – Less use time

when in session– More variable in

use pattern over time

Kaye, et al. 2011 Kaye, et al. Alzheimer & Dementia, 2013

Intact MCI

Continuous, multi-domain assessment over timevia pervasive computing – the future norm…

DataFusion

Improved Assessment & Outcomes

Acknowledgements: The ORCATECH Village

Research Collaborators

Diverse Companies Funding Sponsors

Additional Material

Hayes et al., Proceedings : Engineering in Medicine and Biology Soc, 2006; Leen T, et al., Technology and Aging, 2007 ; Hayes T et al. .Journal of Aging Health, 2009; Hayes T et al.Telemedicine Jounal and E-Health, 2009

Direct Assessment of Everyday Cognition

PROSPECTIVE MEMORY

Prospective memory task – probability of remembering to take medications as desired tracked using a familiar plastic pill box.

vs

Conventional memory task – recall a list of unrelated words.

ORCATECH MedTracker

Medication Adherence: A Sensitive Measure of Cognitive Function

• Adherence assessed with MedTracker taking a vitamin BID; target times set by seniors

• Mean Age - 83 yrs

• Assessed continuously x 5 wks

• Based on ADAScog: Lower Cognition Group (n =18) vs Higher Cognition Group (n = 20)

• Very mild cognitive change in independent elderly is associated with medication adherence

• Medication adherence can be a very early marker of cognitive and functional impairment.

Hayes T et al., Journal of Aging Health, 2009

0

10

20

30

40

50

60

70

80

90

100

Lower Cognition

Higher Cognition

Median time within 12.0 mins of goal

Median time within 53.4 mins of goal

% A

dh

ere

nt

* Significantly worse Adherence in Lower Cognition Group

MotionDetectors

Contact/DoorSwitches

PhoneSensors

Load Cells /Bed Sensors

LocationTracking

MedicationTracker

Computer

Raw Sensor Data

Sleep

Phone Use

Weight

MedicationEvents

DeparturesArrivals

GaitVelocity

ComputerInteractions

WeightScale

Mobility

LocationEstimation

Sleep Hygiene

Socialization

MedicationAdherence

Depression

PhysicalImpairments

Direct Assessment

Inference

Memory

Attention

Information Level Fusion

Sensor Level Fusion

Cognitive Decline

Change Detection

Mirabella Portland, a new generation of Living Laboratories

Weekly on-line reports provides unique insights into

function: patterns of low mood

“During the last week, have you felt downhearted or blue for more than three days?”

(2008-2010)N = 122; 14,566 reports

Seasonal Pattern ofLow Mood Reports

Thielke, unpublished, 2013

Social Engagement RCT Hiroko Dodge, PI

Face-to-Face

Internet email/VOIP

Telephone

Channels of Engagement