Technical Innovation Needed John A. Stankovic BP America - - PowerPoint PPT Presentation

technical innovation needed
SMART_READER_LITE
LIVE PREVIEW

Technical Innovation Needed John A. Stankovic BP America - - PowerPoint PPT Presentation

Technical Innovation Needed John A. Stankovic BP America Professor Department of Computer Science University of Virginia http://www.cs.virginia.edu/wsn/medical/ http://wirelesshealth.virginia.edu/ Panelists Jack Stankovic, UVA Mani


slide-1
SLIDE 1

Technical Innovation Needed

John A. Stankovic BP America Professor Department of Computer Science University of Virginia

http://www.cs.virginia.edu/wsn/medical/

http://wirelesshealth.virginia.edu/

slide-2
SLIDE 2

Panelists

  • Jack Stankovic, UVA
  • Mani Srivastava, UCLA
  • Anind Dey, CMU
slide-3
SLIDE 3

Discussion Questions

  • What are the key open (technical)

research questions for sensing, actuation, and system-integration to support aging in place?

  • What are the limits of current technolgy?
slide-4
SLIDE 4

True Research Partnership

  • Technical solutions informed by real

medical problems

  • Research on two levels
  • Not a service!!!!
slide-5
SLIDE 5

Two Issues

  • Requirements
  • Realisms
slide-6
SLIDE 6

Requirements

  • Accuracy requirement for detecting ADLs?
  • We don’t know!

– Accuracy on toilet visits to detect prostrate problems? – Accuracy on quality of sleep to detect insomnia and/or contribute to detection of depression?

  • Current: accurate as possible
slide-7
SLIDE 7

Realisms

  • Humans and their behaviors are not

simple

  • Example: Sleep
  • Environments are not simple
  • Example: Acoustics
slide-8
SLIDE 8

Realities – Sounds Encountered

Physiological: Sneezing, nose blowing, sniffling, clearing throat, hiccup, eating, burp, humming, laughter, drinking, snoring Objects: phone vibrating or ringing, typing, mouse wheel, unwrapping food, papers rustling, clothes rustling, television, piano, moving furniture, doors

  • pening and closing, objects dropping or moving,

footsteps, pouring liquid, coffee percolation, dishwasher, cleaning sounds Ambient: truck backing up, siren, birds chirping, passing airplane, traffic, motorized tools (lawnmower, etc)

slide-9
SLIDE 9

Main Point

  • Many current solutions work ONLY when

humans and environments are (assumed to be) very constrained

slide-10
SLIDE 10

Realisms

  • Activity Recognition (AR) of ADLs

– Higher accuracy required – Overlapped activities – Across room activities – Many realities (missing data) ….

slide-11
SLIDE 11

Main Point

  • Normal behavior is very complex

– Per day – On Wednesdays – Two times per week – Every other month – In summer when condition X exists – Grouping of activities – Context dependent – …

slide-12
SLIDE 12

Too Many False Alarms

  • Semantic Anomaly Detection

– Sensor Level Anomalies – Activity Level Anomalies

  • Point
  • Context
  • Collective

– Semantics