Home Health Care and Assisted Living

1
Home Health Care and Assisted Living

• Professor John A. Stankovic
• Department of Computer Science
• University of Virginia

2
Themes

• Unobtrusive and wireless sensor devices and
  networks
• Support for many different medical problems
• Individual products
• Complete systems

3
Outline

• Examples of Technology for Medicine
• Home Health Care and Assisted Living Stankovic
  et. al.
• Gait Monitoring Weaver et. al.
• Body Sensor Networks Lach et. al.
• Smart Walker Russell et. al.

4
The Problems

• Home Health Care
• (Large Scale) Assisted Living Facilities

5
Smart Living Space
6
The SEAS Vision

• Flexible targeting of care to a persons health
  condition
• Environmental and Physiological Data
• Longitudinal Studies

7
The SEASMedicine Vision

• Flexible targeting of care to a persons health
  condition
• Stroke, Parkinsons, Diabetes, Dementia,
• Environmental and Physiological Data
• (Define new) Longitudinal Studies

8
With Harvard
9
With Harvard
10
With MARC UVA Medical School
11
SATIRE
With the Univ. of Illinois
12
Other Sensor Data

• Physiological
• Pulse
• SpO2
• ECG
• Blood Pressure
• Weight
• (Dust/Pollen)
• Activities
• Walking
• Sitting
• Falling

13
GaitMate Gait AnalysisMark Williams, MD, and
Alfred Weaver, PhD
Attach accelerometers to ankles and sacrum wire
to data recorder next generation equipment is
wireless
Collect 3D motion data from four sensors as
patient walks down hallway, turns around, and
walks back
Software analyzes waveform and automatically
identifies significant events, e.g. heal strike,
toe-off
Physician analyzes graphs to diagnose or predict
disease (e.g., Parkinsons)
14
Body Sensor Networks for Monitoring and Assessing
Movement DisordersPI John Lach
(jlach_at_virginia.edu)Graduate students Adam
Barth, Mark Hanson, Harry Powell

• Application examples
• Tremor assessment for Parkinsons Disease and
  Essential Tremor study, diagnosis, treatment
• Gait analysis for movement disorder diagnosis and
  fall risk assessment
• Assessing efficacy of Cerebral Palsy physical
  therapy treatments
• Key system metrics
• Wearable (small, light, easy to use)
• Low power (long lifetime with small battery)
• Configurable (system can be adapted for specific
  applications)

Wireless sensor node
Tremor frequency domain analysis example (high
energy at 5Hz reveals tremor)
15
NSF WALKER TEAM
16
(No Transcript)
17
Home Health Care and Assisted Living

• AlarmNet emulated assisted living facility

18
PDA Real-Time Queries
AlarmGate SW on stargate
19
Circadian Rhythms
Circadian activity rhythm per room for 70 days
20
Behavioral Deviation
Diurnal/nocturnal activity
Life Habits at-home
Learning period
21
Summary/Vision

• Tailored to Patient Health
• Stroke, Parkinsons, Diabetes, Incontinence,
• Improve Health Care
• Improve Quality of Life
• Reduce Medical Errors
• Continuous Monitoring
• More natural settings
• More complete
• Collect Data for Longitudinal Studies

22
Summary/Vision

• Unobtrusive body networks (smart clothes)
• Seemlessly integrate into larger wireless sensor
  network
• Combine Environment, Activities and Individual
  Physiological Data
• Provide continuous 24/7 care, if needed and as
  needed
• Detect anomalies and react
• Learn correlations to prevent disease
• Effectiveness of treatment