Physiological monitoring

• US 5,999,846 A
• Filed: 11/08/1996
• Issued: 12/07/1999
• Est. Priority Date: 11/08/1995
• Status: Expired due to Fees

• Current Assignee: Oxford Biosignals (Us) Incorporated
• Original Assignee: Oxford Medical Limited (Oxford Instruments PLC)
• Inventors: Pardey, James, Dadswell, Michael Richard, Laister, Mark Jeremy, Tarassenko, Lionel
• Primary Examiner(s): Nasser, Robert L.
• Application Number:
  US08/745,780
• Time in Patent Office: 1,124 Days
• Field of Search: 600/509, 600/513, 600/544, 600/555, 600/546, 600/545
• US Class Current: 600/544
• CPC Class Codes: A61B 5/369 Electroencephalography [EEG] devices for psychotechnics A61B5/16
• CPC Class Codes: A61B 5/4812 Detecting sleep stages or cycles
• CPC Class Codes: A61B 5/4815 Sleep quality
• CPC Class Codes: A61B 5/6843 Monitoring or controlling sensor contact pressure
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

An insomnia or vigilance monitor comprising one or more electrodes (1a,1b) for obtaining an electrical signal from a subject over a period of epochs, the electrical signal being related to the sleep or wakefulness stage type being experienced by the subject; and a processor (5) adapted to analyze the electrical signal and assign a sleep or wakefulness stage type to each epoch to generate a hypnogram. Methods of monitoring sleep or vigilance using the mastoid site are also disclosed. Further disclosures relate to a method of training and testing a first neural network for use in a physiological monitor, and a method of assigning a class to an epoch of a physiological signal obtained from a subject as a set of samples.

Claims

1. A vigilance monitor comprising:

(1) one or more electrodes for obtaining an electrical signal from a subject over a period of epochs, said electrical signal being related to a wakefulness stage type being experienced by said subject;

(2) a processor adapted to analyze said electrical signal and assign one of a plurality of wakefulness stage types to each of said epochs to generate a Wakeogram;

(3) means for monitoring said Wakeogram to determine whether said Wakeogram meets predetermined criteria; and

(4) means responsive to said means for monitoring said Wakeogram to generate a message when said Wakeogram meets said predetermined criteria.

2. A vigilance monitor according to claim 1, wherein said electrical signal comprises a single electroencephalogram (EEG) channel, and said Wakeogram is generated from said electrical signal only.

3. A vigilance monitor according to claim 1, wherein said electrical signal is obtained from an electrode site which cannot necessarily be scored by humans.

4. A vigilance monitor according to claim 1, further comprising a memory adapted to store said Wakeogram.

5. A vigilance monitor according to claim 1, wherein said monitor comprises a single portable unit.

6. A vigilance monitor according to claim 1, wherein said processor comprises a neural network.

7. A method of training and testing a first neural network for use in a physiological monitor, the method comprising;

(1) obtaining a first set of physiological signals from a subject, said first set being obtained over a first period of epochs on a subject;

(2) obtaining a second set of physiological signals from said subject, said second set being obtained over said first period of epochs, and having a correlation with said first set of physiological signals;

(3) assigning a class to each of said epochs by analyzing said first set of physiological signals by a known method;

(4) separating said first set of physiological signals into a first set of training signals and a first set of test signals;

(5) separating said second set of physiological signals into a second set of training signals and a second set of test signals;

(6) training a second neural network by inputting said first set of training signals and using said classes assigned to said epochs as training labels;

(7) training said first neural network by inputting said second set of training signals and using said classes assigned to said epochs as training labels; and

(8) monitoring said first neural network by comparing a class assigned to each of said epochs by said first and second neural networks when input with said second set of test signals and said first set of test signals to obtain an indication of performance.

8. A method according to claim 7, wherein said first and second sets of physiological signals comprise EEG signals.

9. A method of assigning a class to an epoch of a physiological signal obtained from a subject as a set of samples, the method comprising

(1) estimating a probability of each of a plurality of stage types for each of said samples;

(2) cumulatively multiplying said probabilities for each of said samples with the probabilities of previous samples;

(3) determining which of said stage types has a higher probability than all other of said stage types when all of said samples in said epoch have been cumulatively multiplied; and

(4) assigning said determined stage type to said epoch.

10. A method according to claim 9, wherein said physiological signal comprises an EEG signal.

11. A method of vigilance monitoring, the method comprising:

(1) obtaining an electrical signal from a subject over a period of epochs, said electrical signal being related to a wakefulness stage type being experienced by said subject;

(2) analyzing said electrical signal and assigning one of a plurality of wakefulness stage types to each of said epochs to generate a Wakeogram;

(3) monitoring said Wakeogram to determine whether said Wakeogram meets predetermined criteria; and

(4) generating an alarm message when said Wakeogram meets the predetermined criteria.

12. A method according to claim 11, wherein said electrical signal comprises a single EEG channel, and said Wakeogram is generated from said single EEG channel only.

13. A method according to claim 11, wherein said analyzing of said electrical signal is performed using a neural network.