Adam Hoover, Eric Muth

1
Adam Hoover, Eric Muth

• The Weight Watch

2
Outline
1. Introduction
2. Related work
3. Example result
4. Conclusion
3
Introduction

• The problem

• 26 U.S. adults, ages 20-74, are obese (U.S
  Department of Health and Human Services)
• Related health risks cardiovascular disease,
  diabetes, cancer, hypertension etc.
• Currently used treatments exercise, diet,
  behavior modification, drugs, surgery

4
Introduction

• Real-time caloric intake measurement/feedback

• caloric output can be measured in many exercises
  in real-time caloric intake only measured after
  the fact (after consumption finished)
• caloric intake can be recorded in diary,
  patterns not used in feedback
• caloric intake can be carefully controlled
  through total meal preparation, requires
  controlled eating

5
Introduction

• Our Solution Wrist-band caloric intake monitor

• measures wrist motion to detect bites
• provides vibratory feedback to slow down,
  stop, etc.

6
Introduction

• Benefits

• Private - Wont embarrass people in public
• Remind people in real time, before
  unwanted/unneeded calories taken in

7
Related work

• Related Work Wrist device applications

• Health monitoring ambulatory versus
  non-ambulatory time
• Energy expenditure exercise monitor
• Other virtual keyboard

8
Related work

• Related Work Measuring caloric intake

• All-day video analysis post-review film
• Preeaten vs. post-eaten scale/image
  measurements
• Other food imaging to determine fat content

9
Related work

• Related Work Patents for eating control devices

• Hardwired setup (e.g. plate fork) to detect
  bites
• Esophageal and gastrointestinal tract sensors
  (clinical)
• Throat (Adams apple) scarf sensor
• Arm raise/lower timer-based sensor
• Hoover Muth wrist rotary motion

10
Example a few bites
Yellow true bite, pink false bite
11
Example several minutes
12
Result A Record on Nov 24, 2007
13
Result

• True detected detect a bite when people bite
• Undetected dont detect a bite when people
  bite
• False detected detect a bite when people dont
  bite

14
Conclusion

• Status

• Detect bite in real time
• True detection 80, False detection

• Future Work

• Global analysis find long-term patterns
  (individual)
• Implement warning part build embedded system
• Study use of device in combating obesity and
  eating disorders
• Additional sensors, motion types/patterns
• Calendar/long-term monitoring uses