Method and apparatus for noise reduction of electromyogram signals

• US 20030125635A1
• Filed: 12/27/2001
• Published: 07/03/2003
• Est. Priority Date: 12/27/2001
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Zheng, Yibin, Maalouf, Khalil John
• Granted Patent: US 6,678,551 B2
• Time in Patent Office: Days
• US Class Current: 600/546
• CPC Class Codes: A61B 5/391 of genito-urinary organs
• CPC Class Codes: A61B 5/4356 Assessing uterine contractions A61B5/033 takes precedence
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal

Abstract

A system and a method are provided for reducing noise in Electromyogram (EMG) signals from an external abdominal surface to detect uterine contractions. The system comprises an EMG1 sensor configured to detect a first EMG signal and to generate a corresponding EMG1 input signal and an EMG2 sensor configured to detect a second EMG signal and to generate a corresponding EMG2 input signal. A signal processor is coupled to the EMG1 sensor and the EMG2 sensor. The signal processor processes the EMG1 input signal and the EMG2 input signal with a filter and sum technique to generate an EMG1 reduced noise signal that represents a magnitude of at least one contraction event and periodicity of a set of multiple contraction events.

Claims

1. A system for processing electromyogram (EMG) input signals from an external abdominal surface to detect uterine contractions, said system comprising:

an EMG1 sensor configured to detect a first EMG signal and to generate a corresponding EMG1 input signal;

an EMG2 sensor configured to detect a second EMG signal and to generate a corresponding EMG2 input signal; and

a signal processor coupled to said EMG1 sensor and said EMG2 sensor, said signal processor being configured to process said EMG1 input signal and said EMG2 input signal with a filter and sum technique to generate an EMG1 reduced noise signal that represents a magnitude of at least one contraction event and periodicity of a set of multiple contraction events.

2. The system of claim 1 further comprising:

an envelope detection processor configured to process said EMG1 reduced noise signal to generate an EMG1 display signal representing said magnitude of at least one said contraction event and said periodicity of said set of multiple contraction events.

3. The system of claim 2 further comprising a display device coupled to said envelope detection processor, wherein said display device is configured to display said EMG1 display signal.

4. The system of claim 2, wherein said envelope detection processor is configured to process said EMG1 reduced noise signal by utilizing a envelope detection method including auto regression.

5. The system of claim 1, said signal processor further comprising:

an EMG1 adaptation computation processor;

an EMG1 Finite Impulse Response (FIR) filter; and

an EMG2 reduced noise FIR filter,

wherein said EMG1 adaptation computation processor is configured to operate a constrained power reduction technique to generate weight coefficients for filter taps in said EMG1 FIR filter and in said EMG2 reduced noise FIR filter;

said weight coefficients of said EMG1 FIR filter being selected to preserve a desired signal in said EMG1 reduced noise signal, said weight coefficients of said EMG2 reduced noise FIR filter being selected to reduce output power.

6. The system of claim 5, wherein said EMG1 FIR filter and said EMG2 reduced noise FIR filter have a model order number of at least 2.

7. The system of claim 5, wherein a digitized representation of said EMG1 input signal and a digitized representation of said EMG2 input signal have a sampling frequency range from about 1 Hz to about 200 Hz, and said EMG1 reduced noise signal has a frequency response range from about 0.01 Hz to about 3 Hz.

8. The system of claim 5, wherein said EMG1 FIR filter and said EMG2 reduced noise FIR filter have a model order number of about 21 and a digitized representation of said EMG1 input signal and a digitized representation of said EMG2 input signal have a sampling frequency of about 4 Hz.

9. The system of claim 5, wherein said EMG1 adaptation computation processor is configured to utilize an auto regresion method to determine said weight coefficients of said EMG1 FIR filter and said EMG2 reduced noise FIR filter.

10. The system of claim 9, wherein said auto regression method uses a Least Mean Square adaptation algorithm.

11. The system of claim 9, wherein said auto regression method uses a buffered data technique.

12. The system of claim 1, wherein said signal processor is further configured to process said EMG1 input signal and said EMG2 input signal with said filter and sum technique to generate an EMG2 reduced noise signal and further comprising;

an envelope detection processor, configured to independently process said EMG1 reduced noise signal and said EMG2 reduced noise signal through an envelope detection method to generate an EMG1 display signal and an EMG2 display signal each of which represents a magnitude of at least one contraction event and periodicity of a set of multiple contraction events.

13. The system of claim 12, further comprising;

a display device;

wherein said display device is coupled to said envelope detection processor, wherein said display device is configured to display at least one of said EMG1 display signal and said EMG2 display signal.

14. The system of claim 12, wherein said envelope detection processor is further configured to generate a processed version of said EMG1 reduced noise signal and a processed version of said EMG2 reduced noise signal, wherein said envelope detection processor is further configured to process said processed version of EMG1 reduced noise signal and said processed version of EMG2 reduced noise signal to generate a weighted average processed EMG display signal.

15. The system of claim 14, further comprising:

a display device;

wherein said display device is configured to be coupled to said envelope detection processor, wherein said display device is configured to display at least one of said EMG1 display signal, said EMG2 display signal, and said weighted average processed EMG display signal.

16. The system of claim 12, wherein said envelope detection processor is further configured to generate a processed version of said EMG1 reduced noise signal and a processed version of said EMG2 reduced noise signal, wherein said envelope detection processor is further configured to generate a processed EMG display signal which comprises a highest signal value of said processed version of said EMG1 reduced noise signal and said processed version of said EMG2 reduced noise signal.

17. The system of claim 16, further comprising:

a display device;

wherein said display device is configured to be coupled to said envelope detection processor, wherein said display device is configured to display at least one of said EMG1 display signal, said EMG2 display signal and said processed EMG display signal.

18. The system of claim 1, said signal processor further comprising:

an EMG2 adaptation computation processor;

an EMG1 reduced noise FIR filter; and

an EMG2 FIR filter,

wherein said EMG2 adaptation computation processor is configured to operate a constrained power reduction technique to generate of weight coefficients for filter taps in said EMG1 reduced noise FIR filter and in said EMG2 FIR filter;

said weight coefficients of said EMG2 FIR filter being selected to preserve a desired signal in said EMG2 reduced noise signal, said weight coefficients of said EMG1 reduced noise FIR filter being selected to reduce output power.

19. The system of claim 18, wherein said EMG2 FIR filter and said EMG1 reduced noise FIR filter have a model order number of at least 2.

20. The system of claim 18, wherein a digitized representation of said EMG2 input signal has a sampling frequency range from about 1 Hz to about 200 Hz, and said EMG2 reduced noise signal has a frequency response range from about 0.01 Hz to about 3 Hz.

21. The system of claim 18, wherein said EMG2 FIR filter and said EMG1 reduced noise FIR filter have a model order number of about 21 and a digitized representation of said EMG2 input signal has a sampling frequency of about 4 Hz.

22. The system of claim 18, wherein said EMG2 adaptation computation processor utilizes an auto regresion method to determine said weight coefficients of said EMG2 FIR filter and said EMG1 reduced noise FIR filter.

23. The system of claim 22, wherein said auto regression method uses a Least Mean Square adaptation algorithm.

24. The system of claim 22, wherein said auto regression method uses a buffered data technique.

25. The system of claim 1, wherein further comprising:

an alpha electrode;

a bravo electrode; and

a charlie electrode, wherein said alpha electrode, said bravo electrode and said charlie electrode are configured to be placed in contact with said exterior abdominal surface adjacent to said uterus, wherein said alpha electrode and said bravo electrode are configured to be coupled to said EMG1 sensor to detect said first EMG signal, wherein said bravo electrode and said charlie electrode are configured to be coupled to said EMG2 sensor to detect said second EMG signal.

26. The system of claim 1, wherein further comprising:

an alpha electrode;

a bravo electrode;

a charlie electrode; and

a delta electrode, wherein said alpha electrode, said beta electrode, said charlie electrode and said delta electrode are configured to be placed in contact with said exterior abdominal surface adjacent to said uterus, wherein said alpha electrode and said bravo electrode are configured to be coupled to said EMG1 sensor to detect said first EMG signal, wherein said charlie electrode and said delta electrode are configured to be coupled to said EMG2 sensor to detect said second EMG signal.

27. The system of claim 26, wherein said each of said electrodes is selected from the group consisting of an Electrocardiogram (EKG) electrode and an Electromyogram (EMG) electrode.

28. A method for processing electromyogram (EMG) input signals from an external abdominal surface to detect uterine contractions, said method comprising:

generating an EMG1 input signal;

generating an EMG2 input signal;

processing said EMG1 input signal and said EMG2 input signal with a filter and sum technique to generate an EMG1 reduced noise signal; and

processing said EMG1 reduced noise signal with an envelope detection method to generate an EMG1 display signal, wherein said EMG1 display signal represents a magnitude of at least one contraction event and periodicity of a set of multiple contraction events.

29. The method of claim 28 further comprising the EMG1 display signal on a display device.

30. The method of claim 28, wherein a digitized representation of said EMG1 input signal and a digitized represenation of said EMG2 input signal have a sampling frequency rate range from about 1 Hz to about 200 Hz and said EMG1 reduced noise signal has a frequency response range from about 0.01 Hz to about 3 Hz.

31. The method of claim 28, wherein processing said EMG1 input signal and said EMG2 input signal with said filter and sum technique to generate said EMG1 reduced noise signal further comprises the steps of:

operating a constrained power reduction technique to generate a set of weight coefficients for filter taps in an EMG1 FIR filter and in an EMG2 reduced noise FIR filter by selecting said weight coefficients of said EMG1 FIR filter, wherein a desired signal in said EMG1 reduced noise signal is preserved and selecting said weight coefficients of said EMG2 reduced noise FIR filter so as to reduce an output power.

32. The method of claim 31, wherein said EMG1 FIR filter and said EMG2 reduced noise FIR filter have a model number of at least 2.

33. The method of claim 28 further comprising the steps of:

processing said EMG1 input signal and said EMG2 input signal with a filter and sum technique to generate an EMG2 reduced noise signal; and

processing said EMG2 reduced noise signal with an envelope detection method to generate an EMG2 display signal, wherein said EMG2 display signal represents a magnitude of at least one contraction event and periodicity of a set of multiple contraction events.

34. The method of claim 33 further comprising displaying the EMG2 display signal on a display device.

35. The method of claim 33, wherein said EMG2 reduced noise signal has a frequency response range from about 0.01 Hz to about 3 Hz.

36. The method of claim 33, wherein processing said EMG1 input signal and said EMG2 input signal with said filter and sum technique to generate an EMG2 reduced noise signal further comprises:

operating a constrained power reduction technique to generate a set of weight coefficients for filter taps in an EMG2 FIR filter and in an EMG1 reduced noise FIR filter, selecting said weight coefficients of said EMG2 FIR filter, wherein a desired signal in said EMG2 reduced noise signal is preserved; and

selecting said weight coefficients of said EMG1 reduced noise FIR filter so as to reduce an output power.

37. The method of claim 36, wherein said EMG2 FIR filter and said EMG1 reduced noise FIR filter have a model number of at least 2.

38. The method of claim 33, further comprising:

processing said EMG1 reduced noise signal through an envelope detection method to generate a processed version of said EMG1 reduced noise signal;

processing said EMG2 reduced noise signal through an envelope detection method to generate a processed version of said EMG2 reduced noise signal; and

processing said processed version of EMG1 reduced noise signal and said processed version of EMG2 reduced noise signal to generate a weighted average processed EMG display signal.

39. The method of claim 38 further comprising displaying said weighted average processed EMG display signal on a display device.

40. The method of claim 33, further comprising:

processing said EMG1 reduced noise signal through an envelope detection method to generate a processed version of said EMG1 reduced noise signal;

processing said EMG2 reduced noise signal through an envelope detection method to generate a processed version of said EMG2 reduced noise signal; and

selecting a highest signal value of either said processed version of said EMG1 reduced noise signal or said processed version of said EMG2 reduced noise signal to generate to generate a processed EMG display signal.

41. The method of claim 40, further comprising displaying said processed EMG display signal on a display device.