# US 8,784,324 B2 · Sleep monitoring system > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 8,784,324 B2 - Document ID: US8784324B2 - Status: Granted - Kind: B2 - Application number: 11/718,301 - Filed: 2005-11-02 - Published: 2009-05-21 - Granted: 2014-07-22 - Earliest claimed: 2004-11-02 - Assignees: University College Dublin, National University of Ireland - Applicants: Conor Heneghan; Stephen Redmond - Inventors: Conor Heneghan; Stephen Redmond - Web page: https://patentgrep.com/patent/US8784324B2 ## CPC - A61B 5/4806: {Sleep evaluation (A61B5/4821 takes precedence; devices for inducing sleep A61M21/02)} - A61B 5/02405: {Determining heart rate variability} - A61B 5/113: occurring during breathing - A61B 5/4812: {Detecting sleep stages or cycles} - A61B 5/7264: {Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems} - G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems - A61B 5/7267: {involving training the classification device} - A61B 5/7278: {Artificial waveform generation or derivation, e.g. synthesizing signals from measured signals} - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA ## Abstract A sleep monitoring system includes an ECG device (2) and a respiration inductance plethysmogram (3) which monitor cardiac activity and physical (ribcage) respiration respectively and feed representative signals to a digital data processor. Operations (5-9) process the beat interval data, while in a second thread, operations (20-24) independently process the amplitude modulation of the ECG data caused by the respiratory motion of the subject. The inductance plethysmogram device (3) provides an input to the processor which represents respiration as directly monitored independently of the ECG. Operations (30-34) process this direct respiration data independently and in parallel, in a third thread. All extracted features are fed to a classifier which in step (10) combines selected combinations of features to make decisions in real time. ## Claims ### Claim 1 A sleep monitoring system comprising: an interface for receiving sensor signals; wherein the sensor signals include signals from both electrocardiogram and respiration sensors; a processor for extracting a plurality of features from the sensor signals; and the processor is configured to operate as a classifier for generating an output indicating a sleep stage based upon classification of the plurality of extracted features, wherein at least one of the plurality of extracted features includes a measure of respiration and electrocardiogram derived respiratory (EDR) features, and wherein the classifier uses a combination of the plurality of extracted features derived from frequency-domain and time-domain parameters of the electrocardiogram and respiration sensor signals to provide an estimate of the sleep stage based on probability distributions of the extracted features. ### Claim 2 (depends on 1) A sleep monitoring system as claimed in claim 1, wherein the processor extracts time domain and frequency domain features from the measure of respiration. ### Claim 3 (depends on 1, 2) A sleep monitoring system as claimed in claim 1 or 2, wherein the processor measures heartbeat intervals from cardiogram sensor signals and uses said measurements to extract features. ### Claim 4 (depends on 3) A sleep monitoring system as claimed in claim 3, wherein the extracted features include mean interval per epoch, standard deviation of intervals, longest interval, and shortest interval. ### Claim 5 (depends on 3) A sleep monitoring system as claimed in claim 3, wherein the processor measures amplitude modulation of the cardiogram sensor signals caused by respiratory motion of a person and uses said measurement to extract features. ### Claim 6 (depends on 5) A sleep monitoring system as claimed in claim 5, wherein the extracted features include variance of a derived respiratory signal, and power of the respiratory signal at a frequency band. ### Claim 7 (depends on 5) A sleep monitoring system as claimed in claim 5, wherein the extracted features include the dominant frequency of respiration and the power at the dominant respiratory frequency. ### Claim 8 (depends on 5) A sleep monitoring system as claimed in claim 5, wherein the processor independently extracts features from the heartbeat interval measurements and from the amplitude modulation measurements. ### Claim 9 (depends on 1) A sleep monitoring system as claimed in claim 1, wherein the interface receives sensor signals from a device for physically monitoring patient respiration. ### Claim 10 (depends on 9) A sleep monitoring system as claimed in claim 9, wherein the processor measures variations in signals from said device. ### Claim 11 (depends on 9, 10) A sleep monitoring system as claimed in claim 9 or 10, wherein the processor uses said measurements to independently extract features. ### Claim 12 (depends on 9, 10) A sleep monitoring system as claimed in claim 9 or 10, wherein the features extracted by the processor from the signals from said device include ribcage respiration effort in each of a plurality of frequency bands, envelope power, and breath length variation. ### Claim 13 (depends on 1) A sleep monitoring system as claimed in claim 1, wherein the interface receives sensor signals from devices simultaneously monitoring patient respiration and patient cardiograms, and the processor simultaneously processes said signals. ### Claim 14 (depends on 13) A sleep monitoring system as claimed in claim 13, wherein said device comprises an inductance plethysmograph. ### Claim 15 (depends on 9) A sleep monitoring system as claimed in claim 9, wherein the processor uses measurements from said sensor signals to extract a first set of features and independently uses sensor signals derived from a cardiogram to extract a second set of features. ### Claim 16 (depends on 1) A sleep monitoring system as claimed in claim 1, wherein the features are extracted for each of a series of epochs. ### Claim 17 (depends on 16) A sleep monitoring system as claimed in claim 16, wherein the epochs have a duration of less than 30 seconds. ### Claim 18 (depends on 16) A sleep monitoring system as claimed in claim 16, wherein the processor extracts a plurality of detrended features derived from a plurality of epochs in a sequence. ### Claim 19 (depends on 18) A sleep monitoring system as claimed in claim 18, wherein the detrended features are generated by subtracting a local mean signal from epoch feature values. ### Claim 20 (depends on 1) A sleep monitoring system as claimed in claim 1, wherein the classifier operates according to a discriminant classifier model. ### Claim 21 (depends on 1) A sleep monitoring system as claimed in claim 1, wherein the search process identifies an optimal subset of features to use for classification performance. ### Claim 22 (depends on 21) A sleep monitoring system as claimed in claim 21, wherein the sequential forward floating search process uses a coefficient that is a measure of an inter-rater agreement taking account of a prior probability of a specific class occurring. ### Claim 23 (depends on 22) A sleep monitoring system as claimed in claim 22, wherein said process executes passes which add a feature that most improves performance to already-selected features. ### Claim 24 (depends on 1) A sleep monitoring system as claimed in claim 1, wherein the classifier generates the output indicating sleep stage based upon a subset of the plurality of extracted features.