1. 26/03/2012
Seminar on
“DESIGN AND IMPLEMENTATION OF EMBEDDED
MONITOR SYSTEM FOR DETECTION OF A PATIENT’S
BREATH BY DOUBLE WEBCAMS”
by
ABHISHEK SOMAYAJI
1AP08IS002
8th semester
Department of Information Science and Engineering
APS College Of Engineering

2. Agenda
• Introduction
 Methods
• Operation of the system
 Temporal differencing algorithm
 Breath monitor flowchart
 Implementation results
 Software interface of EMSBD
• Different cases
• Advantage and disadvantage
• Conclusion
• References

3. Introduction
• Traditional way of monitoring a patient’s breath rate
requires contact with the body by tying a device to it.
• For ex,
 Impedance Pneumography
 Respiratory inductive Plethysmography
• Design detects chest expansion and contraction in a
way that is similar to the detection of a moving
object by image processing.
Methods
 Temporal Differencing
 Background Subtraction
 Optical Flow

4. Operation environment of the
system

5. • Two webcams
• Embedded board.
• LCD displays for real time status of breath rate on
Embedded board.
• Embedded board instead of PC
 Low power consumption.
 Low cost.
 Portability.

6. Using temporal differencing
methods to detect micro-vibration

8. • Gaussian Filter removes noise.
• Smoothing filter
• Degree of smoothing depends on standard
deviation.
• Avoids misjudgement.
• The Gaussian distribution function is

10. BREATH
MONITORING
FLOWCHART

11. Abnormalities & Implementation
Result

12. • Integration of embedded board with web cam.
• Uses kernel of ARM 11
• Processing speed is 667 MHz
• 3D graphics accelerator helps to process
images easily.
• Operating system for coding is linux.
• C-code program

13. Software Interface of EMSBD
• Indicators show the direction of movements on LCD display.
• Image data consists of 4 parameter
 Direction – up or down.
 Transition .
Number of breaths.
Breath rate.

14. • Three transitions is equal to one complete
breath cycle.
• Relationship between parameters is shown by
(1) and (2).

15. Breath displacement diagrams
• Demonstrates both chest expansion and
contraction.
• Horizontal axis is time.
• Vertical axis is distance.
• Normal man’s respiratory rate 16 to 20 per
minute.
• Women 18 to 22 per minute.
• Newborn 40 to 44 per minute.

16. • Case 1:
 Recording time 60 secs
 19 full breaths.
 Normal case

17. • Case 2:
 Recording time is 60 secs.
 10 full breaths
 Slow rate but still not harmfull

18. • Case 3:
 Recording time is 60 secs.
 8 full breaths
 Two breathing stops of 10 secs.
 Abnormal.

19. Example of long time
measurement

20. • Sleeping breath rate from 11:00 PM to 7:00
AM.
• Horizontal axis- time in minutes.
• Vertical axis- breath rate
• 480 pieces of information.
• Syndrome when there is continuous 5 secs and
10 secs breathing stops.
• If continues for more than 10 seconds it will
be dangerous.

22. Advantage & Disadvantage
• Advantage
 Avoids Inconvenience of any contact with the body.
 Alarm system is very helpful.
 A warning message is sent to the hospital by means
of the network interface board.
 Protect the patient’s privacy.
• Disadvantage
 It is unable to discover the outline of the moving
object completely.
 Very sensitive to changes in the environment.

23. Conclusion
• Avoids inconvenience of any contact with the
body.
• Alarm system is very helpful.
• Also monitor micro-vibration.

24. References
• [1] Chunhui Zhao, Wei Liu, Yi Wang, Yongmei
Cheng and Hongcai Zhang Language and Image
Processing - ICALIP, pp. 143-146, 7-9 July, 2008.
• [2] Yongseok Yoo and Tae-Suh Park, Proceedings of
the 2008 IEEE International
• Conference on Computer Vision and Pattern
Recognition Workshops CVPR, pp. 1-8, 23-29 June,
2008.
• [3] Dong Wang, Hong Zhu , Qin Li , Yong Chu and
Ruirui Ji, Proceedings of the 2006 IEEE International
Conference on Information Acquisition, pp. 332-336,
20-23 Aug, 2006.