Method and system for analysis of long term physiological polygraphic recordings

• US 5,299,118 A
• Filed: 09/05/1991
• Issued: 03/29/1994
• Est. Priority Date: 06/26/1987
• Status: Expired due to Fees

• Current Assignee: Viasys Healthcare Incorporated (Cardinal Health, Inc.)
• Original Assignee: Nicolet Instrument Corporation (Thermo Fisher Scientific Incorporated)
• Inventors: Martens, Wim L. J., Kap, Jaap I.
• Primary Examiner(s): Envall, Jr., Roy N.
• Assistant Examiner(s): Huntley, David
• Application Number:
  US07/755,164
• Time in Patent Office: 936 Days
• Field of Search: 364/413.01, 364/413.05, 128/731, 395/155
• US Class Current: 600/509
• CPC Class Codes: A61B 5/369 Electroencephalography [EEG] devices for psychotechnics A61B5/16
• CPC Class Codes: A61B 5/372 Analysis of electroencephalograms
• CPC Class Codes: A61B 5/389 Electromyography [EMG]
• CPC Class Codes: A61B 5/398 Electrooculography [EOG], e.g. detecting nystagmus; Electroretinography [ERG]
• CPC Class Codes: A61B 5/4809 Sleep detection, i.e. determining whether a subject is asleep or not
• CPC Class Codes: A61B 5/4812 Detecting sleep stages or cycles
• CPC Class Codes: A61B 5/4818 Sleep apnoea
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• 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: G16H 40/63 for local operation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: Y10S 128/92 Computer assisted medical diagnostics
• CPC Class Codes: Y10S 706/924 Medical

Abstract

An analysis system accepts physiological sensor signals, including electroencephalogram (EEG) and other signals commonly sensed in sleep analysis, stores the raw data for later retrieval, extracts features from the digitized signal data which are of a type apt to be significant in classifying and detecting physiological functions (such as the state of sleep) and matches the extracted features to patterns which indicate the type of feature that has been extracted. The matched features are then utilized to classify for each epoch (limited period of time) the state of the physiological function (such as the stage of sleep) for that epoch. The results of the classification analysis are displayed on a CRT screen to the operator on a real time basis and in time correlation with representations of detected features from the various physiological signals. Upon completion of the tests, the operator can interact with the system to change the patterns by which the features are matched and the classification criteria, and can then re-analyze the data and have it redisplayed to demonstrate the results of the re-analysis.

Claims

1. A system for analyzing physiological signals obtained from physiological sensors n a subject to determine stages of sleep comprising:

(a) means for receiving time varying physiological signals from the sensors on the subject and for digitizing the signals to provide digital physiological signal data;

(b) means for storing the digitized physiological signal data in correlation with time and making such data available for later long term retrieval;

(c) analysis means receiving the digital physiological signal data for initially analyzing the data on a real time basis to detect patterns therein indicating fluctuations in physiological function using criteria, wherein the analysis means analyses the data to determine the physiological function of sleep to produce an analysis result, and wherein the physiological sensor signals include time varying EEG signal data, and wherein the analysis means includes means for separating the time varying EEG signal data into a plurality of signals corresponding to frequency bands generally conforming to the frequencies of standard EEG waveform components, including frequencies for delta waves, the theta waves, beta waves and sleepspindles;

(d) display means including a display screen for displaying to an operator on the display screen the results of the analysis from the analysis means on a real time basis, wherein the display means displays the results of the analysis as a hypnogram of the stages of sleep as a function of time;

(e) input means receiving commands from the operator for providing new criteria specifications from the operator to the analysis means, wherein the analysis means includes means for reanalyzing the physiological data stored for long term retrieval utilizing the new criteria for determining fluctuations in physiological function, wherein the display means includes means for, during reanalysis, displaying to the operator a matrix of emulated classification rules having characterizing parameters listed on one axis of the matrix and classes on the other axis of the matrix, and for displaying reference levels and rules to the operator in matrix elements for the initial analysis, and wherein the input mans includes means for receiving input from the operator to change the reference level or rules indicated in the elements of the matrix to a level or rule desired by the operator, and the analysis means further includes means for analyzing the data utilizing the new reference levels and rules as modified by the operator to carry out the classification.

2. The system of claim 1 wherein the analysis means includes means for adjusting the frequency bands for each of the five EEG waveform components in center frequency, band width, and upper and lower cut-off frequencies in an adaptive manner to best fit presently incoming EEG signal data.

3. A method of analyzing physiological signals obtained from physiological sensors on a subject, comprising the steps of:

(a) receiving time varying physiological signals from the sensors on the subject;

(b) digitizing the physiological signals to provide digital physiological signal data;

(c) storing the digitized physiological signal data in correlation with time and making such data available for later retrieval;

(d) analyzing the physiological signal data according to criteria to detect patterns therein indicating fluctuations in physiological function and classifying a selected state of the subject based on the detected patterns to produce a result;

(e) displaying to an operator on a display screen the results of the analysis including the classification of the state of the subject as a function of time over at least a portion of the time period that the physiological signal data was taken;

(f) simultaneously with the display of results from (e), displaying the physiological signal data as a function of time over a time base which shows a portion of physiological signal data on the screen at any one time to allow the operator to view the physiological signal data;

(g) receiving an input signal from the operator at selected positions in time corresponding to the physiological signal data displayed on the screen to indicate a pattern from a possible group of patterns at a point in time and storing such pattern indication with respect to the point in time of the pattern, including the step of accepting commands from the operator by a cursor displayed on the screen which is controlled by a cursor control device operated by the operator to move the cursor to a bar displaying possible patterns and allowing the operator to determine the pattern of the signal data displayed on the screen by placing the cursor over a selected pattern indicator on the screen and indicating the desired pattern at that point using the cursor control device.

4. The method of claim 3 including the steps of establishing new criteria for the detection of patterns based on the patterns indicated by the operator, storing the new criteria, and analyzing physiological signal data utilizing the new pattern criteria to detect patterns therein indicating fluctuations in physiological function.

5. The method of claim 4 wherein the step of analyzing physiological signal data is done on signal data not previously analyzed.

6. The method of claim 3 wherein the steps of (a) through (g) are repeated on subjects in a class, and including the steps of establishing new criteria for the class of subjects for patterns based on the patterns indicated by the operator, storing the new criteria for the class of subjects, and analyzing physiological signal data utilizing the new pattern criteria to detect patterns therein indicating fluctuations in physiological function.

7. A method of analyzing physiological signals obtained from physiological sensors on a subject, comprising the steps of:

(a) receiving the time varying physiological signals from the sensors on the subject;

(b) digitizing the physiological signals to provide digital physiological signal data;

(c) storing the digitized physiological signal data in correlation with time and making such data available for later retrieval;

(d) analyzing the physiological signal data to detect patterns therein indicating fluctuations in physiological function and classifying a selected state of the subject based on the detected pattern and selected classification criteria to produce a result;

(e) displaying to an operator on a display screen the results of the analysis including the classification of the state of the subject as a function of time over at least a portion of the time period that the physiological signal data was taken;

(f) simultaneously with the display from (e), displaying the physiological signal data as a function of time over a time base which shows a portion of the physiological signal data on the screen at any one time to allow the operator to view the physiological signal data; and

(g) receiving an input signal from the operator at selected positions in time corresponding to the physiological signal data displayed on the screen to indicate a classification of the state of the subject from a possible group of classifications of states at point in time and storing such classification of state with respect to the point in time of the classification, including accepting commands from the operator by a cursor displayed on the screen which is controlled by a cursor control device opeated by the operator to move the cursor to a bar displaying possible classifications and allowing the operator to determine the classification of the data displayed on the screen by placing the cursor over a selected classification indicator on the screen and indicating the desired classification at that point using the cursor control device.

8. The method of claim 7 including the steps of establishing new criteria for classification of states based on the classification indicated by the operator, storing the new criteria, and analyzing physiological signal data utilizing the new classification criteria to detect patterns therein indicating fluctuations in physiological function and classifying a selected state of the subject based on the detected patterns and the new classification criteria.

9. The method of claim 8 wherein the step of analyzing physiological signal data is done on signal data not previously analyzed.

10. The method of claim 7 wherein steps (a) through (g) are repeated on subjects in a class, and including the steps of establishing new criteria for classification of states for the class of subjects based on the classifications indicated by the operator, storing the new criteria for the class of subjects, and analyzing physiological signal data utilizing the new classification criteria to detect patterns therein indicating fluctuations in physiological function and classifying a selected state of the subject based on the detected patterns and the new classification criteria for the class.

11. A method of analyzing physiological signals obtained from physiological sensors on a subject, comprising the steps of:

(a) receiving the time varying physiological signals from the sensors on the subject;

(b) digitizing the physiological signals to provide digital physiological signal data;

(c) storing the digitized physiological signal data in correlation with time and making such data available for later retrieval;

(d) analyzing the physiological signal data to detect patterns therein indicating fluctuations in physiological function and classifying a selected state of the subject based on the detected patterns and selected classification criteria to produce a result;

(e) displaying to an operator on a display screen the results of the analysis including the classification of the state of the subject as a function of time over at least a portion of the time period that the physiological signal data was taken;

(f) simultaneously with the display from (e), displaying the physiological signal data as a function of time over a time base which shows a portion of the physiological signal data on the screen at any one time to allow the operator to view the physiological signal data; and

(g) receiving an input signal from the operator at selected positions in time corresponding to the physiological signal data displayed on the screen to indicate a reclassification of the state of the subject from a possible group of classifications of states at a point in time and storing such classification of state with respect to the point in time of the classification, wherein when the operator has reclassified a particular section of the physiological signal data, the analysis results displayed on the screen which corresponds to the points in time reclassified by the operator are displayed in a different color from the remaining classification to highlight to the operator the points of the classification which were manually classified by the operator.