Fetal data processing system and method employing a time-frequency representation of fetal heart rate

• US 5,957,855 A
• Filed: 05/29/1997
• Issued: 09/28/1999
• Est. Priority Date: 09/21/1994
• Status: Expired due to Fees

• Current Assignee: BETH Israel Deaconess Medical Center (Caregroup Incorporated)
• Original Assignee: BETH Israel Deaconess Medical Center (Caregroup Incorporated)
• Inventors: Bennett, Frederick M., Oriol, Nancy E.
• Primary Examiner(s): Kamm, William E.
• Assistant Examiner(s): SCHAETZLE, KENNEDY
• Application Number:
  US08/809,401
• Time in Patent Office: 852 Days
• Field of Search: 600/301, 600/408, 600/437, 600/453, 600/481, 600/483, 600/500, 600/508, 600/511, 600/528, 600/515, 600/523, 600/586, 600/591, 607/902
• US Class Current: 600/511
• CPC Class Codes: A61B 5/02411 of foetuses
• CPC Class Codes: A61B 5/344 Foetal cardiography
• CPC Class Codes: A61B 5/4035 Evaluating the autonomic nervous system
• CPC Class Codes: A61B 5/4362 Assessing foetal parameters
• 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
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: Y10S 128/925 Neural network

Abstract

A fetal data processing system and method, and a fetal monitor and method for monitoring the condition of a fetus are disclosed. A fetal heart rate time series is received and sampled. A non-linear time-frequency transformation is performed to generate a time-frequency representation of the fetal heart rate time series for heart rate time series data spanning a time period which is preferably less than ten seconds. Analysis of fetal heart rate and fetal heart rate variability and other available data is performed to evaluate fetal well-being. Because of the high time resolution of the transformation, short-term transient variations in the heart rate and heart rate variability are considered in the analysis.

Claims

1. A fetal data processing method comprising:

receiving a signal indicative of fetal heart rate time data for the fetus;

generating a time-frequency representation based on the fetal heart rate time data; and

analysing the time-frequency representation using the fetal heart rate time data to indicate a condition of the fetus.

2. The method of claim 1 further comprising analyzing transient change in fetal heart rate variability.

3. The method of claim 1 further comprising indicating transient change in fetal heart rate.

4. The method of claim 1 wherein the analyzing step further comprises using a feature extraction system to form a feature vector.

5. The method of claim 1 wherein a uterine contraction signal is used to perform the analyzing step.

6. The method of claim 1 wherein a fetal movement signal is used to perform the analyzing step.

7. The method of claim 1 further comprising predicting outcome for the fetus.

8. The method of claim 1 wherein the analyzing step comprises applying a set of classification rules to fetal data.

9. The method of claim 1 wherein the analyzing step comprises applying a pattern of fetal data to a neural network.

10. The method of claim 1 wherein the analyzing step comprises analyzing the time-frequency representation in plural frequency bands.

11. The method of claim 1 wherein the time-frequency representation spans a frequency range from 0.02 Hz to 1.1 Hz.

12. The method of claim 1 wherein the analyzing step comprises computing areas of the time-frequency representation.

13. The method of claim 1 wherein the time-frequency representation is generated for fetal heart rate time data from a time period of less than ten seconds.

14. The method of claim 1 further comprising sampling the fetal heart rate time data at a frequency greater than 2 Hz.

15. The method of claim 1 further comprising displaying in real time variables used to indicate the condition of the fetus.

16. The method of claim 1 further comprising displaying a three-dimensional image of the time-frequency representation.

17. The method of claim 1 further comprising generating recommendations regarding fetal well-being.

18. The method of claim 1 wherein the time-frequency representation is a non-linear time-frequency representation.

19. The method of claim 1 wherein the time-frequency representation is generated using a smoothed Wigner distribution.

20. A fetal data processing system comprising:

an input that receives fetal heart rate time data for a fetus; and

a processing system comprising;

a transformation subsystem that receives the fetal heart rate time data and generates a time-frequency representation for the fetal heart rate time data, and

a variable analysis subsystem that receives the time-frequency representation and fetal condition data and generates a variable representation of fetal condition.

21. The fetal data processing system of claim 20 further comprising a classifying subsystem that analyzes the fetal heart rate time data to indicate a condition of the fetus.

22. The fetal data processing system of claim 21 wherein the classifying subsystem analyzes a uterine contraction signal to indicate the condition of the fetus.

23. The fetal data processing system of claim 21 wherein the classifying subsystem analyzes a fetal movement signal to indicate the condition of the fetus.

24. The fetal data processing system of claim 21 wherein the classifying subsystem analyzes the time-frequency representation in plural frequency bands.

25. The fetal data processing system of claim 21 wherein the classifying subsystem applies a set of classifying rules to a pattern of fetal data to classify the fetal data.

26. The fetal data processing system of claim 21 wherein the classifying subsystem comprises at least one neural network for receiving a pattern of fetal data to classify the fetal data.

27. The fetal data processing system of claim 21 wherein the classifying subsystem comprises a plurality of neural networks that receive patterns of fetal data to classify the fetal data.

28. The fetal data processing system of claim 21 wherein the classifying subsystem predicts an outcome for the fetus.

29. The fetal data processing system of claim 20 further comprising a display for displaying in real time variables used to indicate the condition of the fetus.

30. The fetal data processing system of claim 20 further comprising a display controller for selecting for display in real time variables used to predict outcome for the fetus.

31. The fetal data processing system of claim 20 further comprising a display displaying recommendations regarding fetal well-being.

32. The fetal data processing system of claim 20 further comprising a fetal monitor providing the fetal heart rate time data.

33. The fetal data processing system of claim 20 wherein the transformation subsystem generates the time-frequency representation based on fetal heart rate time data from a time period not greater than 10 seconds.

34. The fetal data processing system of claim 20 further comprising a display for displaying a three-dimensional plot of the time-frequency representation.

35. The fetal data processing system of claim 20 wherein the time-frequency representation generated by the processing system is a non-linear time-frequency representation.

36. A fetal data processing system comprising:

an input that receives fetal heart rate time data for a fetus; and

a processing system comprising;

a transformation subsystem that receives the fetal heart rate time data and generates a time-frequency representation for the fetal heart rate time data; and

a variable analysis subsystem that receives the time-frequency representation and fetal condition data and generates a variable representation of fetal condition; and

a display that displays a three-dimensional plot of the time-frequency representation.