Hydraulic Bed Sensor and System for Non-Invasive Monitoring of Physiological Data

• US 20130197375A1
• Filed: 08/31/2012
• Published: 08/01/2013
• Est. Priority Date: 08/31/2011
• Status: Active Grant

• Current Assignee: Curators of the University of Missouri (University of Missouri)
• Original Assignee: Curators of the University of Missouri (University of Missouri)
• Inventors: Skubic, Marjorie, Heise, David, Rosales, Licet
• Granted Patent: US 11,013,415 B2
• Time in Patent Office: Days
• US Class Current: 600/484
• CPC Class Codes: A61B 2562/0247 Pressure sensors
• CPC Class Codes: A61B 5/0205 Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• CPC Class Codes: A61B 5/024 Detecting, measuring or recording pulse rate or heart rate A61B5/0205, A61B5/021 take precedence
• CPC Class Codes: A61B 5/0816 Measuring devices for examining respiratory frequency measuring frequency of electric signals G01R23/00
• CPC Class Codes: A61B 5/1102 Ballistocardiography
• CPC Class Codes: A61B 5/113 occurring during breathing
• CPC Class Codes: A61B 5/6892 Mats
• CPC Class Codes: A61B 5/72 Signal processing specially adapted for physiological signals or for diagnostic purposes character recognition, recognising digital ink, document-oriented image-based pattern recognition G06V30/00; recognition or understanding of postal images G06V30/424

Abstract

Disclosed herein is a new and improved non-invasive bed sensing system for detecting and monitoring physiological movements such as heartbeat and respiration. The system may employ a hydraulic fluid to transduce the physiological pressures to an integrated pressure sensor and a new and improved signal processing method to identify individual cardiac pulses from the electronic signals generated by the hydraulic transducer. The system provides increased sensitivity capable of capturing quantitative pulse and respiration rates with subtle changes, ability to distinguish between instances of low pulse rate and shallow breathing, and improved comfort over existing system.

Claims

1. A health monitoring system, the system comprising:

a hydraulic sensor for positioning with respect to a living being to non-invasively sense a signal, the signal comprising a heartbeat component indicative of a heartbeat pattern for the living being and a respiratory component indicative of a respiratory pattern for the living being; and

a processor for receiving the sensor signal, the processor configured to analyze data representative of the sensor signal to generate extracted heartbeat data.

2. The system of claim 1 wherein the hydraulic sensor comprises a hydraulic transducer.

3. The system of claim 2 wherein the hydraulic sensor further comprises a pressure sensor, wherein the hydraulic transducer is configured to apply fluid pressure to the pressure sensor in response to the heartbeat and respiratory patterns of the living being.

4. The system of claim 3 wherein the hydraulic transducer comprises a deformable vessel filled with a fluid.

5. The system of claim 4 wherein the deformable vessel comprises a sealed portion, an open portion, and a connector, the connector configured to connect the open portion with the pressure sensor for delivering fluid from the vessel to the pressure sensor to thereby apply the fluid pressure to the pressure sensor.

6. The system of claim 5 wherein the hydraulic sensor is configured to exhibit a flat profile.

7. The system of claim 3 wherein the hydraulic sensor is positioned beneath a mattress for use by the living being in a location along the mattress in substantial alignment to where a torso of the living being is expected when the living being is lying on the mattress.

8. The system of claim 3 further comprising a plurality of the hydraulic sensors.

9. The system of claim 8 wherein the processor is further configured to (1) receive a plurality of sensor signals from the plurality of hydraulic sensors, and (2) analyze data representative of the sensor signals to select which of the sensor signals is to be further analyzed to generate the extracted heartbeat data.

10. The system of claim 8 wherein the hydraulic sensors are positioned beneath a mattress for use by the living being in a plurality of locations along the mattress in substantial alignment with a range of where a torso of the living being is expected when the living being is lying on the mattress.

11. The system of claim 3 wherein the hydraulic sensor is positioned proximate to a mattress, the mattress for use by the living being.

12. The system of claim 11 wherein the hydraulic sensor is positioned beneath the mattress.

13. The system of claim 12 wherein the hydraulic sensor is positioned to contact an underside of the mattress.

14. The system of claim 13 further comprising a memory, wherein the processor is further configured to store the extracted heartbeat data in the memory, the memory comprising a data structure representative of the extracted heartbeat data for the living being over time, and wherein the processor is further configured to interact with the memory to analyze the data structure to determine whether a health condition for the living being is indicated by the data structure.

15. The system of claim 14 wherein the processor is further configured to apply a filtering operation to data representative of the sensor signal to generate extracted respiratory data.

16. The system of claim 15 wherein the processor is further configured to store the extracted respiratory data in the memory, the memory further comprising a data structure representative of the extracted respiratory data for the living being over time, and wherein the processor is further configured to interact with the memory to analyze the heartbeat data structure and the respiratory data structure to determine whether any of a plurality of health conditions for the living being are indicated by the heartbeat data structure and the respiratory data structure.

17. The system of claim 16 wherein the processor is further configured to apply a condition to data representative of the sensor signal to generate extracted restlessness data for the living being, and wherein the processor is further configured to store the extracted restlessness data in the memory, the memory further comprising a data structure representative of the extracted restlessness data for the living being over time, and wherein the processor is further configured to interact with the memory to analyze the heartbeat data structure, the respiratory data structure, and the restlessness data structure to determine whether any of a plurality of health conditions for the living being are indicated by the heartbeat data structure, the respiratory data structure, and the restlessness data structure.

18. The system of claim 14 wherein the processor is further configured to apply a condition to data representative of the sensor signal to generate extracted restlessness data for the living being, and wherein the processor is further configured to store the extracted restlessness data in the memory, the memory further comprising a data structure representative of the extracted restlessness data for the living being over time, and wherein the processor is further configured to interact with the memory to analyze the heartbeat data structure, the respiratory data structure, and the restlessness data structure to determine whether any of a plurality of health conditions for the living being are indicated by the heartbeat data structure, the respiratory data structure, and the restlessness data structure.

19. The system of claim 1 wherein the processor is further configured to apply a windowed peak-to-peak deviation (WPPD) analysis to the data representative of the sensor signal to generate the extracted heartbeat data.

20. The system of claim 19 wherein the processor is further configured to (1) generate a plurality of digital samples that are representative of the sensor signal, and (2) apply the WPPD analysis to the digital samples.

21. The system of claim 20 wherein the processor comprises a signal processing circuit and a computer, the signal processing circuit configured to generate the digital samples, and wherein the computer is configured to apply the WPPD analysis to the digital samples.

22. The system of claim 21 wherein the signal processing circuit comprises an amplifier, a filter, and an analog to digital converter (ADC), wherein the amplifier is configured to amplify the sensor signal, wherein the filter is configured to filter the amplified sensor signal to provide anti-aliasing and noise reduction, and wherein the ADC is configured to sample the filtered amplified sensor signal to generate the digital samples.

23. The system of claim 20 wherein the processor is configured to apply the WPPD analysis by determining a difference between a maximum sample value and a minimum sample value of the samples within a sliding window, the sliding window having a sample size, to thereby generate a plurality of windowed difference values.

24. The system of claim 23 wherein the processor is further configured to perform a low pass filter operation on the windowed difference values to thereby generate a plurality of filtered windowed difference values.

25. The system of claim 22 wherein the processor is further configured to (1) segment the filtered windowed difference values into a plurality of segments, (2) compare the segments with a condition, (3) classify the segments into a plurality of classifications based on the comparison, the classifications comprising a first classification and a second classification, and (4) extract the heartbeat data from the segments having the first classification.

26. The system of claim 25 wherein the condition comprises a threshold value, and wherein the processor is further configured to perform the comparison by (1) determining a maximum filtered windowed difference value within a segment, and (2) comparing the determined maximum filtered windowed difference value with the threshold value, and wherein the processor is further configured to perform the classification by assigning the second classification to the segments whose determined maximum filtered windowed difference value exceeds the threshold value.

27. The system of claim 26 wherein the processor is further configured to perform the classification by assigning the second classification to a window of segments around the each segment whose determined maximum filtered windowed difference value exceeds the threshold value.

28. The system of claim 26 wherein the processor is further configured to generate restlessness data for the living being based on the segments having the second classification.

29. The system of claim 25 wherein the segments comprise a plurality of overlapping segments.

30. The system of claim 24 wherein the processor is further configured to define an adjustable cutoff frequency for the low pass filter operation.

31. The system of claim 30 wherein the processor is further configured to adjust the cutoff frequency on a living being-specific basis.

32. The system of claim 30 wherein the processor is further configured to define an adjustable sample size for the sliding window.

33. The system of claim 32 wherein the processor is further configured to adjust the sample size for the sliding window and the cutoff frequency on a living being-specific basis.

34. The system of claim 20 wherein the processor is further configured to define an adjustable sample size for the sliding window.

35. The system of claim 34 wherein the processor is further configured to adjust the sample size for the sliding window on a living being-specific basis.

36. The system of claim 34 wherein the processor is further configured to gauge a regularity for the extracted heartbeat data and dynamically adjust the sample size for the sliding window in response the gauged regularity.

37. The system of claim 1 wherein the processor is further configured to apply a clustering analysis to the data representative of the sensor signal to generate the extracted heartbeat data.

38. The system of claim 37 wherein the clustering analysis comprises a k-means clustering algorithm.

39. The system of claim 38 wherein the processor is further configured to classify the data representative of the sensor signal into a plurality of clusters, the plurality of clusters comprising a first cluster corresponding to a heartbeat and a second cluster corresponding to a non-heartbeat.

40. The system of claim 38 wherein the processor is further configured to (1) generate a plurality of digital samples that are representative of the sensor signal, (2) extract a plurality of features from a plurality of segments of the digital samples, and (3) apply the k-means clustering algorithm to the extracted features for the segments.

41. The system of claim 40 wherein the processor is further configured to extract the features by computing (1) a first distance, (2) a second distance, and (3) a summation of the first and second distances for each segment, the first distance representing a distance from a reference to a maximum for each segment, and the second distance representing a distance from a reference to a minimum for each segment.

42. The system of claim 41 wherein the processor is further configured to (1) sort the segments into two clusters based on the extracted features for each segment, (2) determine which of the two clusters contains the most segments for a time interval, (3) select the cluster determined to contain the most segments as the cluster containing J-peaks for the heartbeat component, and (4) compute the heartbeat data based on the selected cluster.

43. The system of claim 40 wherein the processor comprises a signal processing circuit and a computer, the signal processing circuit configured to generate the digital samples, and wherein the computer is configured to extract the features and apply the k-means clustering algorithm.

44. The system of claim 43 wherein the signal processing circuit comprises an amplifier, a filter, and an analog to digital converter (ADC), wherein the amplifier is configured to amplify the sensor signal, wherein the filter is configured to filter the amplified sensor signal to provide anti-aliasing and noise reduction, and wherein the ADC is configured to sample the filtered amplified sensor signal to generate the digital samples.

45. The system of claim 40 wherein the processor is further configured to bandpass filter the digital samples prior to application of the k-means clustering algorithm.

46. The system of claim 1 further comprising a sensor network for positioning in a living space of the living being, the sensor network comprising the hydraulic sensor and a plurality of other sensors, the other sensors comprising at least one motion sensor, and wherein the processor is further configured to receive and analyze signals from the other sensors to generate additional data for assessing a health status for the living being.

47. A method comprising:

hydraulically transducing physiological movement of a living being to sense a signal, the signal comprising a heartbeat component indicative of a heartbeat pattern for the living being and a respiratory component indicative of a respiratory pattern for the living being; and

processing, by a processor, the sensed signal, the processing including the processor extracting heartbeat data from data representative of the sensor signal.

48. The method of claim 47 wherein the processing step comprises the processor applying a windowed peak-to-peak deviation (WPPD) analysis to data representative of the sensor signal to generate the extracted heartbeat data.

49. The method of claim 47 wherein the processing step comprises the processor applying a clustering analysis to data representative of the sensor signal to generate the extracted heartbeat data.

50. A health monitoring system, the system comprising:

a hydraulic sensor for positioning with respect to a living being to non-invasively sense a signal, the signal comprising a heartbeat component indicative of a heartbeat pattern for the living being and a respiratory component indicative of a respiratory pattern for the living being; and

a processor for receiving the sensor signal, the processor configured to (1) process the sensor signal to generate heartbeat data and respiratory data for the living being, and (2) store the generated heartbeat and respiratory data in a database for subsequent analysis.

51. The system of claim 50 wherein the processor is further configured to (1) process the sensor signal to generate restlessness data for the living being, and (2) store the generated restlessness data in the database for subsequent analysis.

52. An apparatus comprising:

a processor for receiving a sensor signal, the signal comprising a heartbeat component indicative of a heartbeat pattern for a living being and a respiratory component indicative of a respiratory pattern for the living being, the processor configured to apply a windowed peak-to-peak deviation (WPPD) analysis to data representative of the sensor signal to generate extracted heartbeat data.

53. An apparatus comprising:

a processor for receiving a sensor signal, the signal comprising a heartbeat component indicative of a heartbeat pattern for a living being and a respiratory component indicative of a respiratory pattern for the living being, the processor configured to apply a clustering analysis to data representative of the sensor signal to generate extracted heartbeat data.

54. A computer program product comprising:

a plurality of computer-executable instructions resident on a non-transitory computer-readable storage medium for execution to process data representative of a sensor signal, the signal comprising a heartbeat component indicative of a heartbeat pattern for a living being and a respiratory component indicative of a respiratory pattern for the living being, the instructions configured to analyze the data representative of the sensor signal to generate extracted heartbeat data.

55. The computer program product of claim 54 wherein the instructions are further configured to apply a windowed peak-to-peak deviation (WPPD) analysis to the data representative of the sensor signal to generate the extracted heartbeat data.

56. The computer program product of claim 54 wherein the instructions are further configured to apply a clustering analysis to the data representative of the sensor signal to generate the extracted heartbeat data.

57. A sensor comprising:

a hydraulic transducer for positioning with respect to a living being to non-invasively sense a signal, the signal comprising a heartbeat component indicative of a heartbeat pattern for the living being and a respiratory component indicative of a respiratory pattern for the living being; and

a pressure sensor; and

wherein the hydraulic transducer is configured to apply fluid pressure to the pressure sensor in response to the heartbeat and respiratory patterns of the living being.

58. The sensor of claim 57 further comprising a plurality of the hydraulic transducers and a plurality of the pressure sensors arranged in respective pairs.

59. The sensor of claim 58 further comprising a processor in communication with the pressure sensors, each pressure sensor configured to generate a sensor signal in response to the applied fluid pressure from its paired hydraulic sensor, the processor configured to (1) receive the sensor signals from the pressure sensors, (2) analyze the received sensor signals, and (3) select a sensor signal from among the received sensor signals in response to the analysis.

60. A sensing system for non-invasive detecting/monitoring physiological data of a subject resting in bed, said system comprising:

a sensing network with at least one hydraulic transducer attached with a pressure sensor to detect electronic signals generated by said subject'"'"'s physiological movements;

signal acquisition circuitry for collecting the electronic signals; and

a signal processor with an amplifier/filter hardware subsystem for anti-aliasing and noise reduction to result filtered electronic signals and a signal processing algorithm with a Windowed Peak-to-Peak Deviation function to process said filtered electronic signals and extract physiological data of the subject.

61. The device of claim 60, whereas at least part of said sensing network is placed under a mattress, a pillow or a cushion of said subject.

62. The device of claim 60, whereas said amplifier/filter hardware subsystem further comprising:

an amplification stage; and

a filter stage.

63. A method for extracting physiological respiratory data from a set of electronic signals generated by a subject'"'"'s physiological movement while resting in a bed equipped with a sensing network with a hydraulic transducer, signal acquisition circuitry and a signal processor, said method comprising the steps of:

filtering said electronic signal to a desired frequency;

identifying 1-minute segments without artifacts;

subtracting DC bias from each segment; and

determining respiration rate within each segment by counting peaks detected.

64. A method for extracting physiological heartbeat data from a set of electronic signals generated by a subject'"'"'s physiological movement while resting in a bed equipped with a sensing network with a hydraulic transducer, signal acquisition circuitry and a signal processor, said method comprising the steps of:

segmenting the signal in about 5 to 15-second non-overlapping segments;

validating each segment to reduce signals from artifacts;

filtering signals down to a first desired frequency range, while maintaining a certain number of samples per second, so that relevant characteristics of the heartbeat may be captured;

finding windowed peak-to-peak deviation (WPPD), whereas WPPD may be generated by finding the difference between the most negative and the most positive within the sliding window, the size of which may by adjusted to accommodate the pulse rates;

filtering WPPD down further a second desired frequency range to reduce the effect of noise and smooth the signal; and

determining heart rate within each segment by counting the peaks of the low-pass filtered WPPD within a segment and multiplying by a factor to result per-minute heart rate.

65. The method of claim 64, whereas said first desired frequency range is between 5 to 10 Hz, and said second desired frequency range is between 1 to 4 Hz.