Maternal-fetal monitoring system

• US 20050267377A1
• Filed: 05/27/2005
• Published: 12/01/2005
• Est. Priority Date: 05/28/2004
• Status: Active Grant

• Current Assignee: Convergent Engineering, Inc., University of Florida Research Foundation, Incorporated (State University System of Florida)
• Original Assignee: University of Florida Research Foundation, Incorporated (State University System of Florida)
• Inventors: Euliano, Neil Russell II, Marossero, Dorothee, Principe, Jose C., Euliano, Tammy Y.
• Granted Patent: US 8,275,451 B2
• Time in Patent Office: Days
• US Class Current: 600/511
• CPC Class Codes: A61B 5/0011 Foetal or obstetric data
• CPC Class Codes: A61B 5/02411 of foetuses
• CPC Class Codes: A61B 5/033 Uterine pressure
• CPC Class Codes: A61B 5/344 Foetal cardiography
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle
• CPC Class Codes: A61B 5/391 of genito-urinary organs
• CPC Class Codes: A61B 5/4362 Assessing foetal parameters
• CPC Class Codes: A61B 5/6805 Vests
• CPC Class Codes: A61B 5/6831 Straps, bands or harnesses
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00

Abstract

A maternal-fetal monitoring system for use during all stages of pregnancy, including antepartum and intrapartum stages. The maternal-fetal monitoring system of the subject invention comprises (1) a set of sensors; (2) an amplifying/filtering means; (3) a computing means; and (4) a graphical user interface. Accurate clinical data, which can be extracted and provided to the user in real-time using the system of the invention, include without limitation, maternal electrocardiogram (ECG) signals, maternal uterine activity signals (EHG), maternal heart rate, fetal ECG signals, and fetal heart rate. In a preferred embodiment, the maternal-fetal monitoring system of the invention includes an intelligence means, such as a neural network system, to analyze and interpret clinical data for use in clinical diagnosis antepartum, intrapartum and postpartum, as well as delivery strategy.

Claims

1. A maternal-fetal monitoring system comprising:

a) at least two surface sensors to receive a mixture of maternal and fetal vital signals; and

b) a computing means for analyzing the mixture of maternal and fetal vital signals from the sensors, said computing means comprising hardware and software, wherein said software comprises an independent component analysis function that separates maternal from fetal vital signals in real-time, and wherein said software enables automatic selection of maternal and fetal vital signals separated by the independent component analysis function.

2. The maternal-fetal monitoring system of claim 1, wherein the software further comprises a quality assessing function for establishing the quality of maternal and fetal vital signals separated by the independent component analysis means.

3. The maternal-fetal monitoring system of claim 2, wherein the quality assessing function establishes the quality of maternal and fetal vital signals selected from the group consisting of:

ECG, EHG, and HR.

4. The maternal-fetal monitoring system of claim 2, wherein said software enables automatic selection of best quality maternal and fetal vital signals separated by the independent component analysis function.

5. The maternal-fetal monitoring system of claim 2, wherein the separated maternal and fetal vital signals comprise at least one input variable, and wherein the quality assessment function performs the following tasks:

establishing at least one ideal value for a clean maternal vital signal and fetal vital signal;

performing at least one probability function on each separated maternal vital signal and fetal vital signal; and

comparing at least one input variable to the ideal value(s).

6. The maternal-fetal monitoring system of claim 5, wherein the quality assessing function compares any one or combination of the following input variables to the ideal value(s) for clean maternal and fetal vital signals, said variables selected from the group consisting of:

estimated fetal heart rate;

variance of estimated fetal heart rate;

number of false positives;

number of false negatives;

amplitude of estimated fetal QRS peaks;

ratio of autocorrelation peak to variance of autocorrelation function;

sparsity of estimated fetal ECG signal; and

mutual information.

7. The maternal-fetal monitoring system of claim 5, wherein the probability function(s) is a Gaussian function.

8. The maternal-fetal monitoring system of claim 5, wherein the quality assessing function establishes the quality of maternal and fetal vital signals using the principles of fuzzy logic and membership functions.

9. The maternal-fetal monitoring system of claim 1, wherein the sensors are provided on a mesh or vest.

10. The maternal-fetal monitoring system of claim 1, wherein at least two sensors are provided along a midline of a maternal abdomen.

11. The maternal-fetal monitoring system of claim 1, wherein the software further comprises a means for extracting EHG.

12. The maternal-fetal monitoring system of claim 11, wherein at least two sensors are provided along the midline of the maternal abdomen, wherein the maternal and fetal vital signals from the sensors along the midline of the maternal abdomen are input into the EHG extracting means.

13. The maternal-fetal monitoring system of claim 11, wherein the software further comprises a means for providing an EHG spectrogram, a contraction curve, and an estimation of intra-uterine pressure from the EHG extracting means.

14. The maternal-fetal monitoring system of claim 11, wherein the means for extracting EHG comprises at least one down-sample operator for removing extraneous high-frequency information;

at least one filtering operator to obtain an EHG wave envelope; and

at least one contraction detection operator to detect beginning, end, duration, and amplitude of a uterine muscle contraction.

15. The maternal-fetal monitoring system of claim 11, wherein the means for extracting EHG further comprises at least one intelligence system.

16. The maternal-fetal monitoring system of claim 15, wherein the intelligence system is selected from the group consisting of:

artificial neural networks, fuzzy logic, evolutionary computation, knowledge-based systems, optimal linear or nonlinear filters, and artificial intelligence.

17. The maternal-fetal monitoring system of claim 1, further comprising an amplifier for amplifying maternal and fetal vital signs extracted by the sensors.

18. The maternal-fetal monitoring system of claim 1, wherein the software further comprises a means for detecting QRS complex peaks in the maternal and fetal vital signals.

19. The maternal-fetal monitoring system of claim 18, wherein the means for detecting QRS complex peaks comprises a band-pass filter;

a differentiation function; and

a moving average function.

20. The maternal-fetal monitoring system of claim 19, wherein the means for detecting QRS complex peaks comprises a butterworth band-pass filter, a differentiation function of 5 point derivatives, and a moving average function of a 20 point window.

21. The maternal-fetal monitoring system of claim 18, wherein the means for detecting QRS complex peaks in maternal and fetal vital signals provides clinical data selected from the group consisting of:

RR intervals, instantaneous heart rate, beat to beat heart rate, average heart rate, heart rate variance, and estimated false positive and false negative QRS complexes.

22. The maternal-fetal monitoring system of claim 1, wherein the independent component analysis means is selected from the group consisting of blind source separation operations;

Infomax ICA operations;

minimum mutual information operations;

maximum entropy and minimum mutual information operations; and

Mermaid ICA operations.

23. The maternal-fetal monitoring system of claim 1, wherein the independent component analysis means analyzes maternal and fetal vital signals to provide clinical data selected from the group consisting of estimated fetal ECG and estimated maternal ECG.

24. The maternal-fetal monitoring system of claim 1, wherein the software further comprises an auto-correlation means for processing the separated maternal and fetal vital signals.

25. The maternal-fetal monitoring system of claim 1, wherein the software further comprises a means for automatically determining the location of fetal vital signals and maternal vital signals.

26. The maternal-fetal monitoring system of claim 1, wherein the system is portable.

27. A maternal-fetal monitoring system comprising:

a) at least two surface sensors to receive a mixture of maternal and fetal vital signals; and

b) a computing means for analyzing the mixture of maternal and fetal vital signals from the sensors, said computing means including hardware and software, wherein said software comprises a means for estimating intra-uterine pressure.

28. The maternal-fetal monitoring system of claim 27, wherein said software further comprises an EHG extraction operation;

wherein said EHG extraction operation comprises at least one down-sample operator;

at least one filtering operator; and

at least one contraction detection operator to provide at least one EHG extraction operation output; and

wherein said means for estimating intra-uterine pressure utilizes said EHG extraction operation output to estimate intra-uterine pressure.

29. The maternal-fetal monitoring system of claim 28, further comprising a neural network system.

30. A maternal-fetal monitoring system comprising:

a) at least two surface sensors to receive a mixture of maternal and fetal vital signals; and

b) a computing means for analyzing the mixture of maternal and fetal vital signals from the sensors, said computing means including hardware and software, wherein said software comprises a means for calculating montevideo units.

31. The maternal-fetal monitoring system of claim 30, further comprising a neural network system.

32. A maternal-fetal monitoring system comprising:

a) at least two surface sensors for placement on a maternal abdomen to receive a mixture of maternal and fetal vital signals; and

b) a computing means for analyzing the mixture of maternal and fetal vital signals from the sensors, said computing means including hardware and software, wherein said software comprises an independent component analysis function to separate maternal from fetal vital signals in real-time and provide output data; and

a means for reconstructing estimated maternal and fetal vital signals based on said output data.

33. The maternal-fetal monitoring system of claim 32, said software further comprising an input-output matching means for comparing the output data of the independent component analysis function against the mixture of maternal and fetal vital signals from the sensors for use in providing a quality for each sensor placement on the maternal abdomen.

34. The maternal-fetal monitoring system of claim 33, wherein the input-output matching means is selected from the group consisting of:

correlation coefficient analysis, mutual information metric analysis, template matching, or non-linear matching methodology.

35. The maternal-fetal monitoring system of claim 33, wherein the quality for each sensor placement is used by the computing means to provide instructions to the user to reposition the sensor.

36. The maternal-fetal monitoring system of claim 33, wherein the input-output matching means for the sensor is used by the computing means to determine whether to switch the mixture of maternal and fetal vital signals from the sensor into or out of the independent component analysis function.

37. The maternal-fetal monitoring system of claim 32, wherein the estimated maternal and fetal vital signals are ECG, and wherein the reconstructing means comprises at least one filter.

38. The maternal-fetal monitoring system of claim 37, wherein the reconstructing means comprises an inverse filter and a band pass filter.

39. A maternal-fetal monitoring system comprising:

a) at least two surface sensors to receive a mixture of maternal and fetal vital signals; and

b) a computing means for analyzing the mixture of maternal and fetal vital signals from the sensors, said computing means including hardware and software, wherein said software comprises a fetal presentation function and at least one model fetal heart vector corresponding to a fetal presentation, wherein the fetal presentation function performs a comparison analysis of fetal vital signals to the model fetal heart vector to obtain a fetal presentation.

40. The maternal-fetal monitoring system of claim 39, wherein the comparison analysis consists of one or more of the following:

correlation coefficient, mutual information metric, template matching, or non-linear matching methodology.

41. A maternal-fetal monitoring system comprising:

a) at least two surface sensors to receive a mixture of maternal and fetal vital signals; and

b) a computing means for analyzing the mixture of maternal and fetal vital signals from the sensors, said computing means including hardware and software, wherein said software comprises a means for extracting EHG to provide at least one EHG spectrogram having at least one characteristic.

42. The maternal-fetal monitoring system of claim 41, wherein said characteristic is associated with a feature selected from the group consisting of contraction efficiency, preterm labor, probability of vaginal delivery, time to delivery, arrested descent or labor dystocia, the effectiveness of drugs used to induce labor, uterine rupture, labor progress, and post-delivery maternal uterine tone.

43. The maternal-fetal monitoring system of claim 41, wherein said software means further comprises an analyzing means for identifying in the EHG spectrogram said special characteristic and for providing said analysis with regard to said feature to the user.

44. The maternal-fetal monitoring system of claim 41, further comprising a sensor to receive abdominal muscle signals;

wherein said software means further comprises a biofeedback means for directing maternal pushing efforts during labor.

45. A method for monitoring maternal-fetal vital signals from a patient and fetus, comprising:

a) placing at least two surface sensors on the patient'"'"'s abdomen to receive a mixture of maternal and fetal vital signals;

b) connecting the sensors to a computing means for analyzing the mixture of maternal and fetal vital signals from the sensors, said computing means comprising hardware and software, wherein said software comprises an independent component analysis function to separate maternal from fetal vital signals in real-time, and wherein said software enables automatic selection of maternal and fetal vital signals separated by the independent component analysis function; and

c) extracting from the software clinically relevant data with regard to the patient and fetus.

46. The method of claim 45, wherein the software further comprises a quality assessing function for establishing the quality of maternal and fetal vital signals separated by the independent component analysis means.

47. The method of claim 46, wherein the quality assessing function establishes the quality of maternal and fetal vital signals selected from the group consisting of:

ECG, EHG, and HR.

48. The method of claim 46, wherein the separated maternal and fetal vital signals comprise at least one input variable, and wherein the quality assessment function performs the following tasks:

establishing at least one ideal value for a clean maternal vital signal and fetal vital signal;

performing at least one probability function on each separated maternal vital signal and fetal vital signal; and

comparing at least one input variable to the ideal value(s).

49. The method of claim 48, wherein the quality assessing function compares any one or combination of the following input variables to the ideal value(s) for clean maternal and fetal vital signals, said variables selected from the group consisting of:

estimated fetal heart rate;

variance of estimated fetal heart rate;

number of false positives;

number of false negatives;

amplitude of estimated fetal QRS peaks;

ratio of autocorrelation peak to variance of autocorrelation function;

sparsity of estimated fetal ECG signal; and

mutual information.

50. The method of claim 48, wherein the probability function(s) is a Gaussian function.

51. The method of claim 48, wherein the quality assessing function establishes the quality of maternal and fetal vital signals using the principles of fuzzy logic and membership functions.

52. The method of claim 45, wherein the sensors are provided on a mesh or vest.

53. The method of claim 45, wherein the software further comprises an EHG extraction operation, said method further comprising the step of extracting from the software EHG data provided by the EHG extraction operation.

54. The method of claim 53, wherein at least two sensors are provided along the midline of the maternal abdomen, said method further comprising the steps of receiving the maternal and fetal vital signals from the sensors along the midline of the maternal abdomen and providing the maternal and fetal vital signals from the sensors along the midline of the maternal abdomen to the computing means.

55. The method of claim 53, wherein the computing means further comprises a QRS complex peak detection function, said method further comprising the step of controlling the computing means to perform QRS complex peak detection function in the maternal and fetal vital signals.

56. The method of claim 55, wherein the clinically relevant data extracted from the computer means is selected from the group consisting of:

RR intervals, beat to beat heart rate, average heart rate, heart rate variance, and estimated false positive and false negative QRS complexes.

57. The method of claim 45, further comprising the step of controlling the computing means to perform independent component analysis function to extract clinical data selected from the group consisting of estimated fetal ECG and estimated maternal ECG.

58. The method of claim 45, wherein said computing means further comprises an autocorrelating function, said method further comprising the step of controlling the computing means to perform an autocorrelating function on the separated maternal and fetal vital signals from the independent component analysis means.

59. The method of claim 45, wherein said computing means further comprises an operation for determining the location of fetal vital signals channels and maternal vital signals channels, said method further comprising the step of controlling the computing means to perform the operation for determining the location of fetal vital signals channels and maternal vital signals channels.

60. The method of claim 45, wherein the step of extracting clinically relevant data with regard to the patient and fetus comprises extracting variability in fetal heart rate.

61. The method of claim 60, further comprising the step of diagnosing fetal arrhythmias ante-partum using the variability in fetal heart rate.

62. A method for monitoring maternal-fetal vital signals from a patient and fetus, comprising:

a) placing at least two surface sensors on the patient'"'"'s abdomen to receive a mixture of maternal and fetal vital signals;

b) connecting the sensors to a computing means for analyzing the mixture of maternal and fetal vital signals from the sensors, said computing means including hardware and software, wherein said software comprises an EHG extraction operation;

c) extracting EHG data from the mixture of maternal and fetal vital signals from the sensors using the EHG extraction operation;

d) providing information selected from the group consisting of an EHG spectrogram, a contraction curve, and an estimation of intra-uterine pressure based on the extracted EHG data; and

e) extracting from the software clinically relevant data with regard to the patient and fetus.

63. The method of claim 62, wherein the step of extracting EHG data comprises removing extraneous high-frequency information;

obtaining an EHG wave envelope; and

detecting a beginning, end, duration, and amplitude of a uterine muscle contraction.

64. The method of claim 62, wherein the step of extracting from the software clinically relevant data further comprises estimating intra-uterine pressure;

said method further comprising the step of determining failure to progress in labor based on the estimate of intra-uterine pressure.

65. The method of claim 62, wherein the step of extracting from the software clinically relevant data comprises estimating intra-uterine pressure;

said method further comprising the step of determining contraction efficiency.

66. The method of claim 62, wherein the step of extracting from the software clinically relevant data comprises the steps of analyzing the EHG spectrogram and identifying preterm labor.

67. The method of claim 62, wherein the step of extracting from the software clinically relevant data comprises the steps of analyzing the EHG spectrogram and determining post-term muscle health.

68. The method of claim 62, further comprising the steps of placing at least one sensor on the abdomen to receive abdominal muscle signals;

analyzing the EHG spectrogram; and

directing maternal pushing efforts during labor.

69. A method for monitoring maternal-fetal vital signals from a patient and fetus, comprising:

a) placing at least two surface sensors on the patient'"'"'s abdomen to receive a mixture of maternal and fetal vital signals;

b) connecting the sensors to a computing means for analyzing the mixture of maternal and fetal vital signals from the sensors, said computing means including hardware and software, wherein said software comprises an independent component analysis function to separate maternal from fetal vital signals in real-time and provide output data; and

an estimated maternal and fetal vital signals reconstruction operation; and

c) extracting from the software estimated maternal and fetal vital signals using the reconstruction operation.

70. The method of claim 69, wherein the software further comprises an input-output matching means, said method further comprising the steps of:

comparing the estimated maternal and fetal vital signals against the mixture of maternal and fetal vital signals from the at least one sensor using the input-output matching means;

determining a correlation coefficient for the at least one sensor based on the comparison; and

determining a quality of sensor placement based on the correlation coefficient for the at least one sensor.

71. The method of claim 70, further comprising the steps of providing instructions to the user to reposition the corresponding sensor.

72. The method of claim 69, wherein the software further comprises an input-output matching means, said method further comprising the steps of:

comparing the estimated maternal and fetal vital signals against at least one template using the input-output matching means; and

determining a quality of sensor placement based on the results of the comparison.

73. The method of claim 72, further comprising the steps of providing instructions to the user to reposition the corresponding sensor.

74. The method of claim 69, wherein the software further comprises an input-output matching means, said method further comprising the steps of:

comparing the estimated maternal and fetal vital signals against the mixture of maternal and fetal vital signals from at least one sensor using the input-output matching means;

performing mutual information metric analysis; and

determining a quality of sensor placement based on the results of the mutual information metric analysis.

75. The method of claim 74, further comprising the steps of providing instructions to the user to reposition the corresponding sensor.

76. The method of claim 69, wherein the software further comprises an input-output matching means, said method further comprising the steps of:

comparing the estimated maternal and fetal vital signals against the mixture of maternal and fetal vital signals from at least one sensor using the input-output matching means;

performing non-linear matching analysis; and

determining a quality of sensor placement based on the results of the non-linear matching analysis.

77. The method of claim 76, further comprising the steps of providing instructions to the user to reposition the corresponding sensor.

78. The method of claim 69, wherein the maternal and fetal vital signals provided using the reconstruction operation are maternal and fetal ECG, and wherein the reconstruction operation comprises at least one filter.

79. A method for monitoring maternal-fetal vital signals from a patient and fetus, comprising:

a) placing at least two surface sensors on the patient'"'"'s abdomen to receive a mixture of maternal and fetal vital signals;

b) connecting the sensors to a computing means for analyzing the mixture of maternal and fetal vital signals from the sensors, said computing means including hardware and software, wherein said software comprises an intra-uterine pressure estimation operation; and

c) extracting from the software an estimated intra-uterine pressure.