Method and apparatus for assessment of sleep disorders

• US 8,905,928 B2
• Filed: 07/16/2010
• Issued: 12/09/2014
• Est. Priority Date: 07/17/2009
• Status: Active Grant

• Current Assignee: Oregon Health & Science University
• Original Assignee: Oregon Health & Science University
• Inventors: Hayes, Tamara, Hagen, Chad, Pavel, Misha, Beattie, Zachary
• Primary Examiner(s): Thomson, William
• Assistant Examiner(s): SANDS, DAVIN K
• Application Number:
  US13/384,386
• Publication Number:
  US 20120116187A1
• Time in Patent Office: 1,607 Days
• Field of Search: None
• US Class Current: 600/301
• 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 5/0826 Detecting or evaluating apnoea events A61B5/4818 takes precedence
• CPC Class Codes: A61B 5/11 Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb for measuring pulse A61B5/02 ;A61B5/1038 takes precedence; motion detection to correct for motion artifacts in physiological signals A61B5/721
• CPC Class Codes: A61B 5/1102 Ballistocardiography
• CPC Class Codes: A61B 5/113 occurring during breathing
• CPC Class Codes: A61B 5/4812 Detecting sleep stages or cycles
• CPC Class Codes: A61B 5/4818 Sleep apnoea
• CPC Class Codes: A61B 5/6887 mounted on external non-worn devices, e.g. non-medical devices
• CPC Class Codes: A61B 5/6891 Furniture
• 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

Abstract

Embodiments provide systems, methods and apparatuses for monitoring the sleep of a subject in a home environment. In embodiments, load cells placed under bed supports may be coupled to a computing device that may process the load cell data to detect disordered breathing. In some embodiments, a computing device may apply a pattern recognition algorithm to load cell data to distinguish between normal movements and movements associated with a sleep disorder. In an embodiment, apparatuses and methods for monitoring sleep may perform functions associated with detection of sleep disturbances and/or identify a sleep disorder.

Claims

1. A method of differentiating between central apnea, hypopnea, and normal breathing during sleep in a subject, the method comprising:

collecting, by a computing device, load cell signal data from one or more load cells, the load cells being positioned below one or more supports of a bed such that the bed and the one or more bed supports are physically supported by the load cells, the load cell signal data indicating force exerted against the load cell, the collecting being performed while the subject is sleeping;

decimating, by the computing device, the load cell signal data to obtain downsampled load cell signal data;

filtering, by the computing device, the load cell signal data to obtain filtered load cell signal data;

extracting, by the computing device, features from the downsampled load cell signal data;

extracting, by the computing device, features from the filtered load cell signal data;

applying, by the computing device, a classifier to the features extracted from the downsampled load cell signal data and the features extracted from the filtered load cell signal data to classify the load cell signal data, the classifier differentiating between central apnea, hypopnea, and normal breathing; and

identifying, by the computing device, central apnea, hypopnea or normal breathing in the subject based on the classification of the load cell signal data.

2. The method of claim 1, further comprising segmenting, by the computing device, the data into periods of movement and stillness.

3. The method of claim 1, wherein the load cell signal data includes load cell signal data representing periods of stillness.

9. The method of claim 1, wherein extracting features from the downsampled load cell signal data comprises:

calculating a variance in the downsampled load cell signal data;

calculating a normalized average power of the downsampled load cell signal data in a first frequency band;

calculating a normalized average power of the downsampled load cell signal data in a second frequency band greater than the first frequency band;

calculating a normalized average power of the downsampled load cell signal data in a third frequency band greater than the second frequency band; and

calculating a spectral entropy of the down-sampled load cell signal data; and

wherein extracting features from the filtered load cell signal data comprises;

calculating a variance in the filtered load cell signal data;

calculating a range of the filtered load cell signal data; and

calculating a respiration amplitude from the filtered load cell signal data.

10. The method of claim 9, wherein the first frequency band comprises frequencies in an range of 0 Hz to 0.5 Hz, the second frequency band comprises frequencies in an approximate range of 0.6 Hz to 0.75 Hz, and the third frequency band comprises frequencies in an approximate range of 0.75 Hz to 5 Hz.

11. The method of claim 9, wherein for each frequency band, [F₁, F₂], calculating the normalized average power of the downsampled load cell signal data in the frequency band comprises:

multiplying the downsampled load cell signal data by a Blackman window to obtain a modified signal x_B(t);

calculating a power spectral density, S(e^jω

), according to the equation;

12. The method of claim 9, wherein

calculating the respiration amplitude from the filtered load cell signal data comprises:

dividing the filtered load cell signal data into overlapping 5 second windows that have starting points offset by 0.1 seconds;

for each window of the overlapping 5 second windows, calculating a range of the filtered load cell signal data in the window; and

calculating a median of all the ranges.

13. The method of claim 1, wherein the load cells are placed beneath supports of the bed and tensioned to provide support and measure the load at a specific location.

14. The method of claim 9, wherein calculating the spectral entropy of the down-sampled load cell signal data comprises:

calculating a power spectral density of the downsampled load cell signal data;

segmenting the power spectral density into 100 equal frequency bins, f;

normalizing an average power of each frequency bin to obtain a probability density, p_f, for each frequency bin; and

calculating the spectral entropy, SE, according to the equation;

4. An apparatus configured to receive information related to a sleep disorder in a subject, the apparatus comprising:

one or more load cells configured for placement below one or more supports of a bed such that the bed and the one or more bed supports are physically supported by the load cells, the load cells further configured to convert force to an electrical signal indicative of the force; and

a computing device coupled to at least one of the one or more load cells, the computing device comprising computer executable instructions for;

receiving load cell signals from at least one of the one or more load cells;

separately decimating the load cell signals and filtering the load cell signals to obtain downsampled load cell signal data and filtered load cell signal data, respectively;

calculating a variance in the downsampled load cell signal data;

calculating a normalized average power of the downsampled load cell signal data in a first frequency band;

calculating a normalized average power of the downsampled load cell signal data in a second frequency band greater than the first frequency band;

calculating a normalized average power of the downsampled load cell signal data in a third frequency band greater than the second frequency band;

calculating a spectral entropy of the down-sampled load cell signal data;

calculating a variance in the filtered load cell signal data;

calculating a range of the filtered load cell signal data;

calculating a respiration amplitude from the filtered load cell signal data; and

applying a classifier to classify the load cell signals based on the variance in the downsampled load cell signal data, the normalized average power of the downsampled load cell signal data in the first frequency band, the normalized average power of the downsampled load cell signal data in the second frequency band, the normalized average power of the downsampled load cell signal data in the third frequency band, the spectral entropy of the down-sampled load cell signal data, the variance in the filtered load cell signal data, the range of the filtered load cell signal data, and the respiration amplitude from the filtered load cell signal data, the classifier differentiating between central apnea, hypopnea, and normal breathing; and

identifying central apnea, hypopnea or normal breathing in the subject based on the classification of the load cell signal data.

5. The apparatus of claim 4, wherein the classifier is a Bayesian classifier in which parameters of the classifier are approximated using maximum likelihood probabilities of a training set of data which contains examples of sleep patterns of central apnea, hypopnea, and normal breathing.

6. The apparatus of claim 4, further including a transceiver coupled to the computing device or to at least one of the one or more load cells.

7. The apparatus of claim 4, further comprising an alarm, wherein the identifying central apnea, hpopnea, or normal breathing in the subject comprises identifying central apnea or hypopnea in the subject, and wherein the computer executable instructions are operable to actuate the alarm in response to the identifying central apnea or hypopnea in the subject.

8. The apparatus of claim 4, further including a support for a bed, the support configured to be added to the bed and tensioned for use with the one or more load cells.