Multimodal sensing for Pediatric Application

1
Multimodal sensing for Pediatric Application

• M. Annavaram,N. Medvidovic, U. Mitra, S.
  Narayanan G. Sukhatme,
• Z. Meng, S. Qiu, R. Kumar, G. Thatte, D.
  Spruijt-Met
• University of southern California
• Presented By Kartik J Doshi

2
Outline

• Introduction
• KNOWME Network Architecture
• Activity Modeling
• Observations and Ongoing work

3
Introduction

• What?
• Childhood obesity.
• Why?
• Pediatric obesity is a major national and
  international health crisis.
• 2003-2006 11.3 - Age 12-19- obese.
• 16 - Overweight.
• 32 - risk of being
  overwieght.

4
Introduction

• Areas of concentration
• Unified design and evaluation of multimodal
  sensing and interpretation.
• So that decision making and data interpretation
  that will reside within the WBNA allow for
  interactions with the WBAN wearer.

5
KNOWME Network Architecture
6
KNOWME Network Architecture
7
KNOWME Network Architecture

• Sensor system
• What they were Looking for?
• Insight of metabolic activities.
• Capability of data transfer over Bluetooth
• Which sensors they used?
• ECG, pulse oximeter, Body media WMS,
• Accelerometers.
• Nokia N95

8
Activity Modeling

• State of Interest
• Focus on ECG and Accelerometer data.
• Pre- processing of ECG Data.
• Walking, Running Gaussian
• Lying Down, Sitting and standing Laplacian.

9
Activity Modeling
10
Activity Modeling
11
Activity Modeling
12
Observations and Ongoing Work

• Limitations
• Mobile Phone.
• Bluetooth Link.
• Battery power.
• High Activity and Mobility
• Ongoing Work
• Contextual cues
• Power Minimization

13
Critique

• How Energy Expenditure and obesity are related
  and its Impact?

14
Questions ???