US20220313155A1 - Flow-based sleep stage determination - Google Patents
Flow-based sleep stage determination Download PDFInfo
- Publication number
- US20220313155A1 US20220313155A1 US17/653,255 US202217653255A US2022313155A1 US 20220313155 A1 US20220313155 A1 US 20220313155A1 US 202217653255 A US202217653255 A US 202217653255A US 2022313155 A1 US2022313155 A1 US 2022313155A1
- Authority
- US
- United States
- Prior art keywords
- sleep
- user
- breath
- respiratory
- flow signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000008667 sleep stage Effects 0.000 title claims abstract description 60
- 230000000241 respiratory effect Effects 0.000 claims abstract description 92
- 238000000034 method Methods 0.000 claims abstract description 53
- 238000005259 measurement Methods 0.000 claims description 93
- 230000007958 sleep Effects 0.000 claims description 39
- 238000004422 calculation algorithm Methods 0.000 claims description 15
- 230000004622 sleep time Effects 0.000 claims description 10
- 206010062519 Poor quality sleep Diseases 0.000 claims description 9
- 230000029058 respiratory gaseous exchange Effects 0.000 claims description 9
- 208000010340 Sleep Deprivation Diseases 0.000 claims description 4
- 230000004044 response Effects 0.000 claims description 4
- 238000010801 machine learning Methods 0.000 claims description 3
- 230000003860 sleep quality Effects 0.000 claims description 3
- 230000037007 arousal Effects 0.000 claims description 2
- 230000004461 rapid eye movement Effects 0.000 claims 8
- 210000002345 respiratory system Anatomy 0.000 claims 3
- 230000002035 prolonged effect Effects 0.000 claims 1
- 238000002644 respiratory therapy Methods 0.000 claims 1
- 230000001133 acceleration Effects 0.000 description 32
- 230000006870 function Effects 0.000 description 20
- 238000013507 mapping Methods 0.000 description 19
- 230000015654 memory Effects 0.000 description 15
- 239000007789 gas Substances 0.000 description 9
- 238000012545 processing Methods 0.000 description 8
- 238000000605 extraction Methods 0.000 description 7
- 239000011159 matrix material Substances 0.000 description 6
- 238000012360 testing method Methods 0.000 description 6
- 238000012549 training Methods 0.000 description 6
- 230000001225 therapeutic effect Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000003066 decision tree Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000001413 cellular effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000007596 consolidation process Methods 0.000 description 1
- 238000011513 continuous positive airway pressure therapy Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000010348 incorporation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003387 muscular Effects 0.000 description 1
- 208000001797 obstructive sleep apnea Diseases 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000008685 targeting Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4815—Sleep quality
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4812—Detecting sleep stages or cycles
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
- A61B5/087—Measuring breath flow
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4809—Sleep detection, i.e. determining whether a subject is asleep or not
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7225—Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7239—Details of waveform analysis using differentiation including higher order derivatives
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7278—Artificial waveform generation or derivation, e.g. synthesising signals from measured signals
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/74—Details of notification to user or communication with user or patient ; user input means
- A61B5/742—Details of notification to user or communication with user or patient ; user input means using visual displays
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/74—Details of notification to user or communication with user or patient ; user input means
- A61B5/7475—User input or interface means, e.g. keyboard, pointing device, joystick
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M16/00—Devices for influencing the respiratory system of patients by gas treatment, e.g. mouth-to-mouth respiration; Tracheal tubes
- A61M16/021—Devices for influencing the respiratory system of patients by gas treatment, e.g. mouth-to-mouth respiration; Tracheal tubes operated by electrical means
- A61M16/022—Control means therefor
- A61M16/024—Control means therefor including calculation means, e.g. using a processor
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M16/00—Devices for influencing the respiratory system of patients by gas treatment, e.g. mouth-to-mouth respiration; Tracheal tubes
- A61M16/10—Preparation of respiratory gases or vapours
- A61M16/14—Preparation of respiratory gases or vapours by mixing different fluids, one of them being in a liquid phase
- A61M16/16—Devices to humidify the respiration air
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M2205/00—General characteristics of the apparatus
- A61M2205/50—General characteristics of the apparatus with microprocessors or computers
- A61M2205/502—User interfaces, e.g. screens or keyboards
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M2230/00—Measuring parameters of the user
- A61M2230/40—Respiratory characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Definitions
- the invention relates to a method and system for sleep stage determination. More particularly the invention relates to sleep stage determination based at least partly on respiratory signal.
- Devices or systems for providing a humidified gases flow to a user for therapeutic purposes are well known in the art. Such devices are typically configured to detect Sleep Disordered Breathing (SDB) events. Such devices are not usually configured to obtain more detailed information about a user's sleep patterns. One example is the determination of sleep stage of a user.
- SDB Sleep Disordered Breathing
- Sleep stage is determined by the analysis of bio-signals including at least (electroencephalogram (eeg), electromyogram (emg), and electro-oculogram (eog).
- the definition of sleep stages is based on the characteristics of the bio-signals.
- a method of determining a sleep stage of a user comprises receiving a respiratory flow signal of a user; obtaining at least one respiratory feature from at least part of the respiratory flow signal; and determining a sleep stage from the at least one respiratory feature.
- the at least one respiratory feature is based at least partly on at least one duration measurement.
- the at least one duration measurement includes one or more of breath duration, inspiration duration, maximum inspiration time, maximum expiration time, a function of maximum expiration time and maximum inspiration time, a function of inspiration duration and breath duration.
- the at least one respiratory feature is based at least partly on at least one amplitude measurement.
- the at least one amplitude measurement includes one or more of maximum inspiration amplitude, maximum expiration amplitude, a function of maximum inspiration amplitude and maximum expiration amplitude.
- the at least one respiratory feature is based at least partly on at least one centre of mass related measurement.
- the at least one centre of mass related measurement includes one or more of inspiration centre of mass time, inspiration centre of mass amplitude, expiration centre of mass time, expiration centre of mass amplitude, a function of expiration centre of mass time and inspiration centre of mass time, a function of expiration centre of mass amplitude and inspiration centre of mass amplitude.
- the at least one respiratory feature is based at least partly on at least one derivative related measurement.
- the at least one derivative related measurement includes one or more of maximum negative acceleration time, maximum negative acceleration amplitude, maximum positive acceleration time, maximum positive acceleration amplitude, maximum negative flow rate time, maximum negative flow rate amplitude, maximum inspiration acceleration time, maximum inspiration acceleration amplitude.
- the at least one respiratory feature is based at least partly on at least one volume related measurement.
- the at least one volume related measurement includes one or more of inspiration volume, expiration volume, a function of inspiration volume and expiration volume.
- the method further comprises identifying, within the respiratory flow signal, at least one breath signal representing a breath of the user; and obtaining at least one breath measurement from a portion of the respiratory flow signal within which the at least one breath signal is identified.
- the method further comprises identifying, within the respiratory flow signal, a window containing a plurality of breath signals; and obtaining respective breath measurements of the breath signals within the window.
- the at least one breath feature comprises a mean and/or standard deviation of the breath measurements within at least part of the window.
- the method further comprises determining a sleep stage from the at least one breath feature at least partly by applying at least one of a supervised learning algorithm, an unsupervised learning algorithm, a semi-supervised learning algorithm.
- the at least one supervised learning algorithm comprises one or more of linear and logistic regression, support vector machine, artificial neural network, decision tree.
- the sleep stage comprises one of awake, N1, N2, N3, REM.
- the sleep stage comprises one of awake, light sleep, deep sleep, REM.
- the sleep stage comprises one of awake, non-REM, REM.
- the sleep stage comprises one of awake, sleep.
- the method further comprises applying a filter to the respiratory flow signal to remove at least one of high frequency noise, DC level.
- a sleep determination system comprises a feature extractor configured to obtain at least one respiratory feature from at least part of a respiratory flow signal of a user; and a mapping module configured to determine a sleep stage from the at least one respiratory feature.
- a sleep determination system comprises a processor; and a computer readable medium having stored thereon computer executable instructions that, when executed by the processor, cause the processor to perform a method of determining a sleep stage of a user.
- the method comprises receiving a respiratory flow signal of a user; obtaining at least one respiratory feature from at least part of the respiratory flow signal; and determining a sleep stage from the at least one respiratory feature.
- a computer readable medium has stored thereon computer-executable instructions that, when executed by a processor, cause the processor to perform a method of determining a sleep stage of a user.
- the method comprises receiving a respiratory flow signal of a user; obtaining at least one respiratory feature from at least part of the respiratory flow signal; and determining a sleep stage from the at least one respiratory feature.
- the invention in one aspect comprises several steps.
- the relation of one or more of such steps with respect to each of the others, the apparatus embodying features of construction, and combinations of elements and arrangement of parts that are adapted to affect such steps, are all exemplified in the following detailed disclosure.
- connection to includes all direct or indirect types of communication, including wired and wireless, via a cellular network, via a data bus, or any other computer structure. It is envisaged that they may be intervening elements between the connected integers. Variants such as ‘in communication with’, ‘joined to’, and ‘attached to’ are to be interpreted in a similar manner. Related terms such as ‘connecting’ and ‘in connection with’ are to be interpreted in the same manner.
- computer-readable medium should be taken to include a single medium or multiple media. Examples of multiple media include a centralised or distributed database and/or associated caches. These multiple media store the one or more sets of computer executable instructions.
- computer readable medium should also be taken to include any medium that is capable of storing, encoding or carrying a set of instructions for execution by a processor and that cause the processor to perform any one or more of the methods described above.
- the computer-readable medium is also capable of storing, encoding or carrying data structures used by or associated with these sets of instructions.
- computer-readable medium includes solid-state memories, optical media and magnetic media.
- FIG. 1 is a perspective view of a Continuous Positive Airway Pressure (CPAP) machine configured to obtain a respiratory flow signal from a user;
- CPAP Continuous Positive Airway Pressure
- FIG. 2 shows a schematic view of the CPAP machine of FIG. 1 ;
- FIG. 3 shows a schematic view of a sleep stage determination module
- FIG. 4 shows an example of a method performed by the sleep stage determination module of FIG. 3 ;
- FIG. 5 shows examples of duration measurements and amplitude measurements from which the feature extractor of FIG. 3 obtains respiratory feature(s).
- FIG. 6 shows examples of centre of mass measurements from which the feature extractor obtains respiratory feature(s).
- FIG. 7 shows examples of derivative related measurements from which the feature extractor obtains respiratory feature(s).
- FIG. 8 shows examples of volume related measurements from which the feature extractor obtains respiratory feature(s).
- FIG. 9 shows an example of the feature extractor obtaining respiratory feature(s) from window(s) comprising multiple breath signals.
- FIG. 10 shows an example of agreement between the mapping module of FIG. 3 and the inferred sleep stages of a test study.
- FIG. 11 shows an example function of the context module from FIG. 3 applying contextual clues in an attempt to improve accuracy of the performance of the mapping module.
- FIG. 12 shows the accuracy per patient of a test study.
- FIG. 13 shows an example of typical performance for an individual patient.
- FIG. 14 shows an example of a computing system and/or device that implements the sleep stage determination module of FIG. 3 .
- FIG. 1 shows a user 100 receiving air from a modular assisted breathing unit and humidifier system 102 .
- System 102 is shown in FIG. 1 as a Continuous Positive Airway Pressure (CPAP) machine.
- CPAP Continuous Positive Airway Pressure
- the system 102 provides a pressurised stream of heated, humidified gases to the user 100 for therapeutic purposes.
- These therapeutic purposes include for example one or more of reducing the incidence of obstructive sleep apnea, providing CPAP therapy, providing humidification for therapeutic purposes.
- a heated and humidified gases stream passes along the length of a delivery conduit 104 and are provided to the user 100 via a user interface 106 .
- the conduit 104 is heated via a heater wire (not shown) or similar to help prevent rain-out.
- the conduit 104 typically has a circular internal cross-section.
- the internal diameter of the conduit is typically about 20 mm. in an embodiment the internal diameter is between 10 mm and 30 mm. These typical dimensions apply to both flexible portions of the gases flow passageway and rigid components such as elbows and connectors and portions integrated into components of the humidified gases supply.
- the user interface 106 comprises a full face mask. It will be appreciated that alternatives to a full face mask include a nasal mask that surrounds and covers the nose of the user 100 , nasal cannula, tracheostomy fitting, any other suitable user interface.
- FIG. 2 shows a schematic view of system 100 of FIG. 1 .
- the system 100 includes an air supply 200 configured to deliver the heated and humidified gases stream into the delivery conduit 104 .
- the air supply 200 includes a flow control mechanism that in turn includes for example an internal compressor unit and a flow generator or fan unit 202 .
- Air from the atmosphere enters the housing of the air supply 200 via an atmospheric inlet (not shown) and is drawn through the fan unit 202 .
- the output of the fan unit 202 is adjustable by varying the speed of the fan.
- the air supply 200 includes or is interfaced to a humidifier chamber (not shown).
- the humidifier chamber contains a volume of water.
- a heater plate heats the base of the humidifier chamber thereby heating the contents of the chamber. As the water in the chamber is heated it evaporates and the gases within the humidifier chamber become heated and humidified. The gases stream entering the humidifier chamber passes over the heated water and becomes heated and humidified as it does so.
- the heated and humidified gases stream enters the delivery conduit 104 .
- a central controller 204 or control system is connected to a memory 206 .
- the controller 204 receives user input signals via user input 208 and displays output to a user via display 210 .
- the controller 204 receives input from at least one sensor 212 located at various points within system 102 , conduit 104 and/or interface 106 .
- the controller 204 determines a control output that sends signals to adjust the power to the air supply 202 .
- the senor(s) 212 receive signals representative of inspiration and expiration of the user 100 .
- the sensor(s) 212 sends a respiratory flow signal to the controller 204 .
- FIG. 3 shows a schematic view of a sleep stage determination module 300 .
- the module 300 receives a respiratory flow signal 302 .
- An example of a respiratory flow signal is one sent from the sensor(s) 212 to the controller 204 .
- a filter 304 pre-processes the respiratory flow signal to remove high frequency noise and DC level.
- the use of the filter 304 is optional.
- the filter 304 is not present or the respiratory flow signal 302 is not sent to the filter 304 .
- a flow segmenter 306 segments the respiratory flow signal into time-based epochs and/or breath-number-based epochs for subsequent feature extraction as will be further described below.
- the use of the flow segmenter 306 is optional.
- feature extraction is performed on the raw respiratory flow signal 302 .
- a flow segment measurement module 308 determines measurements of the respiratory flow signal 302 on which feature extraction is subsequently based as will be described below. In an embodiment the measurement module 308 determines one or more of duration measurements, amplitude measurements, centre of mass related measurements, derivative related measurements, volume related measurements.
- a feature extractor 310 obtains at least one respiratory feature from the respiratory flow signal 302 .
- the feature extractor takes as input one or more of a filtered respiratory flow signal, a segmented respiratory flow signal in which breath segmentation has been performed, an otherwise unprocessed respiratory flow signal.
- mapping module 312 maps a set of respiratory features to at least one sleep stage as will be further described below.
- a context module 314 optionally applies context data in an attempt to improve the accuracy of the mapping module 312 output.
- the module 300 outputs a sleep stage 316 determined by the mapping module 312 and optionally context module 314 from the respiratory flow signal 302 .
- the module determines a sequence of and/or multiple sleep stages 316 from a respiratory flow signal 302 .
- the module 300 is implemented within the modular assisted breathing unit and humidifier system 102 .
- the module outputs the sleep stage 316 on the display 210 and/or stores the sleep stage on the memory 206 .
- the module 300 is implemented on a computing device for example the computing device 1400 of FIG. 14 .
- the sleep stage 316 is presented on a display and/or maintained on a memory that either form(s) part of the computing device 1400 or is/are connected to computing device 1400 .
- FIG. 4 shows an example of a method 400 performed by the sleep stage determination module 300 of FIG. 3 .
- the module 300 receives 402 a respiratory flow signal. Filtering (not shown) is optionally performed on the respiratory flow signal to remove high frequency noise and DC level.
- the flow segmenter 306 of FIG. 3 optionally applies 404 flow segmentation to the respiratory flow signal. As described above, the flow segmenter segments the respiratory flow signal into time-based epochs and/or breath-number-based epochs for subsequent feature extraction.
- the measurement module 308 obtains 406 flow segment measurements of the respiratory flow signal on which feature extraction is subsequently based.
- measurements include one or more of duration measurements, amplitude measurements, centre of mass related measurements, derivative related measurements, volume related measurements. In an embodiment these measurements comprise breath measurements.
- the feature extractor 310 performs 408 feature extraction on flow segment measurements. For example, the feature extractor 310 obtains at least one respiratory feature from the respiratory flow signal.
- the module 300 determines 410 a sleep stage, a sequence of sleep stages, and/or multiple sleep stages from the features extracted from the respiratory flow signal. In an embodiment this function is performed by the mapping module 312 and optionally also the context module 314 .
- the feature extractor obtains at least one respiratory feature based at least partly on at least one duration measurement.
- FIG. 5 shows examples of duration measurements associated to a maximum inspiration 500 and a maximum expiration 502 .
- the duration measurements include one or more of:
- the duration measurements include a function of two or more duration measurements. Examples of functions include:
- the feature extractor obtains at least one respiratory feature based at least partly on at least one amplitude measurement.
- FIG. 5 also shows examples of amplitude measurements.
- the amplitude measurements include one or more of:
- the amplitude measurements include a function of two or more amplitude measurements.
- An example of a function is a difference between maximum inspiration amplitude and maximum expiration amplitude.
- duration and/or amplitude measurements are obtained from a single breath. In an embodiment the duration and/or amplitude measurements are calculated as a mean or average of at least two breaths. In an embodiment the at least two breaths are consecutive.
- the feature extractor obtains at least one respiratory feature based at least partly on at least one centre of mass measurement.
- FIG. 6 shows examples of centre of mass measurements.
- the centre of mass measurement includes inspiration centre of mass time.
- the centre of mass of the body formed by the inspiration curve is indicated at 600 .
- a time axis component of the inspiration centre of mass 600 is calculated for example relative to the beginning of the breath by the following:
- N Insp is the number of inspiration points i in the curve
- y i and x i are the signal amplitudes and time, respectively.
- the centre of mass measurement includes expiration centre of mass time.
- the centre of mass of the body formed by the inspiration curve is indicated at 602 .
- a time axis component of the expiration centre of mass 602 is calculated for example relative to the beginning of the breath by the following:
- N Exp is the number of expiration points i in the curve
- y i and x i are the signal amplitude and time respectively.
- the centre of mass measurement includes a function of expiration centre of mass time and inspiration centre of mass time.
- An example is the difference between expiration centre of mass time and inspiration centre of mass time.
- the centre of mass measurement includes inspiration centre of mass amplitude.
- An amplitude axis component of the inspiration centre of mass 600 is calculated for example relative to the inspiration curve and the baseline by the following:
- N Insp is the number of inspiration points i in the curve
- y i and x i are the signal amplitude and time, respectively.
- the centre of mass measurement includes expiration centre of mass amplitude.
- An amplitude axis component of the expiration centre of mass 602 is calculated for example relative to the expiration curve and the baseline by the following:
- N Exp is the number of expiration points i in the curve
- y i and x i are the signal amplitude and time, respectively.
- the centre of mass measurement includes a function of expiration centre of mass amplitude and inspiration centre of mass amplitude.
- An example is the difference between expiration centre of mass amplitude and inspiration centre of mass amplitude.
- the feature extractor obtains at least one respiratory feature based at least partly on at least one derivative related measurement.
- FIG. 7 shows examples of derivative related measurements.
- the derivative related measurement includes a maximum negative acceleration time.
- FIG. 7 shows a maximum inspiration 700 and a maximum expiration 702 .
- a maximum negative acceleration 704 has a time value between the time of maximum inspiration 700 and the time of maximum expiration 702 .
- the time value of 704 is a time when the acceleration reaches a negative peak, calculated from the beginning of the breath.
- the time value is computed with the second derivative obtained from a Savitzky Golay filter.
- the derivative related measurement includes a maximum positive acceleration time, having a time value between the time of maximum inspiration 700 and the time of maximum expiration 702 .
- the time value of 706 is a time when the acceleration reaches a positive peak, calculated from the beginning of the breath.
- the time value of 706 is computed with the second derivative obtained from a Savitzky Golay filter.
- the derivative related measurement includes a time value for a maximum inspiration acceleration 710 .
- a time value for maximum inspiration acceleration 710 is a time value between the start of the breath and maximum inspiration 700 when the acceleration is at a maximum.
- the time value is computed with the second derivative obtained from a Savitzky Golay filter.
- the derivative related measurement includes a function of a maximum negative acceleration time and a maximum positive acceleration time.
- a maximum negative flow rate time 708 comprising a time between maximum negative acceleration and maximum positive acceleration when the acceleration is zero, calculated from the beginning of the breath. In an embodiment this is computed by finding a first zero crossing position on a second derivative signal generated by a Savitsky Golay filter.
- the derivative related measurement includes a maximum negative acceleration amplitude.
- the amplitude of the maximum negative acceleration 704 occurs between a time of maximum inspiration 700 and a time of maximum expiration 702 .
- the amplitude of maximum negative acceleration 704 is computed with the second derivative obtained from a Savitzky Golay filter.
- the derivative related measurement includes a maximum positive acceleration amplitude.
- the amplitude of the maximum positive acceleration 706 occurs between a time of maximum negative acceleration 704 and the end of the breath.
- the amplitude of maximum positive acceleration 706 is computed with the second derivative obtained from a Savitzky Golay filter.
- the derivative related measurement includes an amplitude for the maximum inspiration acceleration 710 .
- the amplitude occurs between the start of the breath and maximum inspiration 700 .
- the amplitude is computed with the second derivative obtained from a Savitzky Golay filter.
- the derivative related measurement includes a function of a maximum negative acceleration time and a maximum positive acceleration time.
- An example is an amplitude of maximum negative flow rate 708 comprising an amplitude of a maximum rate of change in the flow signal that occurs between maximum negative acceleration and maximum positive acceleration.
- the amplitude is approximated by calculating the difference in flow at the point when the acceleration is zero.
- the feature extractor obtains at least one respiratory feature based at least partly on at least one volume related measurement.
- FIG. 8 shows examples of volume related measurements.
- the volume related measurement includes inspiration volume 800 comprising the volume of the inspiration part of the flow signal above the baseline.
- inspiration volume 800 is calculated for example by the following:
- N Insp is the number of inspiration points i in the curve
- y i and B are the signal amplitude and the baseline, respectively.
- SR is the sampling rate of the signal in samples per minute.
- the volume related measurement includes expiration volume 802 comprising the volume of the inspiration part of the flow signal above the baseline.
- the expiration volume 802 is calculated for example by the following:
- N Exp is the number of expiration points i in the curve
- y i and B are the signal amplitude and the baseline, respectively.
- SR is the sampling rate of the signal in samples per minute.
- the volume related measurement includes a function of inspiration volume and expiration volume.
- An example is a ratio between inspiration volume and expiration volume.
- FIG. 9 shows an example of feature extraction that in an embodiment is performed by the feature extractor 310 of FIG. 3 .
- a window 900 comprises a plurality of breath signals within a respiratory flow signal. Window 900 is shown centred on breath B indicated at 902 .
- the window is of size N so that it includes N consecutive breaths immediately following breath B in the respiratory flow signal, and includes N consecutive breaths immediately preceding breath B in the respiratory flow signal.
- the boundaries of window 900 are defined so as to include breath +N indicated at 904 , and breath ⁇ N indicated at 906 .
- the value of N is 15. In an embodiment the value of N is selected from a range of 2 to 30.
- window 900 is further subdivided into at least one further window, for example window 908 and window 910 .
- window 908 is 2 ⁇ 3 the size of window 900 .
- window 910 is 1 ⁇ 3 the size of window 900 .
- window 908 and/or window 910 has/have boundaries defined so as to include breath B and breath +N but not breath ⁇ N . In an embodiment, window 908 and/or window 910 has/have boundaries defined so as to include breath B and breath ⁇ N but not breath +N . In an embodiment, window 908 and/or window 910 has/have boundaries defined so as to include breath B but not breath +N and not breath ⁇ N .
- One example of a respiratory feature obtained from at least part of the respiratory flow signal is a mean breath measurement associated to at least some of the breaths in the respiratory flow signal.
- the respiratory features obtained from the respiratory flow signal represent a mean breath measurement associated to the breaths within one or more of window 900 , window 908 , window 910 .
- a respiratory feature obtained from at least part of the respiratory flow signal is a standard deviation of breath measurements associated to at least some of the breaths in the respiratory flow signal.
- the respiratory features obtained from the respiratory flow signal represent a standard deviation of breath measurements associated to the breaths within one or more of window 900 , window 908 , window 910 .
- the breath measurements include one or more of duration measurements, amplitude measurements, centre of mass related measurements, derivative related measurements, volume related measurements.
- the breath features describe the morphology of part of the respiratory flow signal that surrounds a breath B under evaluation.
- the mean and/or standard deviation for up to 26 breath measurements are determined for one or more of windows 900 , 908 , 910 . This results in a total of up to 156 respiratory features.
- a mapping is performed by the mapping module 312 of FIG. 3 .
- the mapping module 312 is configured to map a given set of respiratory features to a sleep stage.
- mapping module 312 applies a supervised learning algorithm. Suitable algorithms include one or more of linear and logistic regression, support vector machine, artificial neural network, decision tree. In an embodiment, the learning is performed offline with previously recorded and labelled data.
- mapping module 312 applies an unsupervised learning algorithm. In an embodiment the mapping module 312 applies a semi-supervised learning algorithm.
- machine learning methods that provide such learning capabilities are used to improve accuracy of the module 300 .
- learning is carried out to suit the conditions of individual patients. For instance, parts of the signal where strong markers are present, for example start of the night or periods prior to events, can be used to adapt the model to specific patient profiles.
- the mapping module 312 maps a set of respiratory features to a sleep stage selected from awake, N1, N2, N3, REM.
- the sleep stage is selected from awake, light sleep, deep sleep, REM.
- the sleep stage is selected from awake, non-REM, REM.
- the sleep stage is selected from awake, sleep.
- the features are calculated on a breath basis.
- the module 300 provides breath-basis output.
- the breath outputs are then combined in 30 second epochs to calculate agreement to a correct hypnogram determined by mapping module 312 .
- the resultant epoch output for each class is calculated as the average of breaths output.
- the resultant epoch output is then normalized so that the sum of the outputs for all classes are 1.
- a context module 314 is configured to apply contextual clues in an attempt to improve accuracy of the mapping module 312 .
- One example of a contextual clue is identification of periods without breathing. It can be assumed that a user is awake after long periods in which the absence of a respiratory flow signal is detected. This occurs in the beginning of the session and when the user takes the mask off during the night. An example threshold is 7 minutes. After this threshold period of no respiratory flow signal it is assumed the user is awake and a sleep stage of awake is determined.
- SDB Sleep Disordered Breathing
- Another example of a contextual clue is neighbouring consensus. It is assumed that individual epochs with differing labels compared to their neighbouring epochs are wrongly detected. One aim is to remove sleep stages that are likely to have been wrongly detected, by analysing the context where they are placed.
- the modelling of sleep stages is performed using algorithms that learn temporal structures at different scales.
- Hidden Markov Models HMM
- TDNN Temporal Delay Neural Networks
- HMM is applied in an attempt to learn the sleep stage transition probabilities using several bio-signals.
- ‘a priori’ probability is implemented to capture probabilities of observing a particular state at different periods of the night.
- the module 300 implements a sleep score.
- the sleep score attempts to capture, in at least one index, sleep quality and effectiveness. In an embodiment, multiple indices are used.
- a good night of sleep should ideally have long periods of deep sleep and REM stages and low sleep fragmentation.
- module 300 calculates one or more of the indices Sleep time (ST), Total Time at Deep Sleep (DST), Total Time at REM (REMT), Number of Sleep Disruptions (number of awakes during the night after sleep onset) (NSD), Number of SDB events (NSDB).
- ST Total Time at Deep Sleep
- REMT Total Time at REM
- NSD Number of Sleep Disruptions (number of awakes during the night after sleep onset)
- NSDB Number of SDB events
- DST score changes linearly [0 for 0 hours, 100 for 1.5 hours or more]
- REMT score changes linearly [0 for 0 hours, 100 for 1.5 hours or more)
- NSD score changes linearly [100 for 0, 0 for 10 or more disruptions]
- NSDB score changes linearly [100 for 0, 0 for 10 or more events/hour].
- the use of the cost learning matrix aims at reducing errors between specific classes.
- the aim is to direct the learning procedure to penalise the misclassification between Awake and REM.
- the operating point is chosen targeting equal TP Rate for all classes.
- the results of the cross-validated test sets have been used to find the equal TP Rate operating points. Having equal TP Rate in all classes implies that similar numbers of samples are allocated to each class when compared to the results of the mapping module 312 .
- FIG. 10 shows an example of agreement between the mapping module 312 and the inferred sleep stages on a test study.
- FIG. 11 shows an application of the context module 314 to improve the accuracy of the mapping module 312 .
- FIG. 13 shows an example of typical performance for an individual patient with agreement of 61.4%.
- FIG. 14 shows an example of a system representative of a computing system and/or device that implements module 300 from FIG. 3 .
- the computing device 1400 comprises one or more of a server of a service provider, a device associated with the client (for example a client device), an on-chip system, any other suitable computing device or computing system.
- computing device 1400 includes a processing system 1402 , one or more computer-readable media 1404 , and one or more Input/Output (I/O) Interfaces 1406 that are communicatively coupled, one to another.
- the computing device 1400 further includes a system bus or other data and command transfer system (not shown) that couples the various components, one to another.
- a system bus includes one or more of a memory bus or memory controller, a peripheral bus, a universal serial bus, a processor or local bus that utilizes any of a variety of bus architectures.
- the processing system 1402 is representative of functionality to perform one or more operations using hardware. Accordingly, the processing system 1402 is illustrated as including hardware element(s) 1408 configured as one or more of processors and/or functional blocks. This may include implementation in hardware as an application specific integrated circuit or other logic device formed using one or more semiconductors.
- the hardware elements 1408 are not limited by the materials from which they are formed or the processing mechanisms employed therein.
- processors may be comprised of semiconductor(s) and/or transistors (e.g., electronic integrated circuits (ICs)).
- processor-executable instructions may be electronically-executable instructions
- the computer-readable media 1404 is illustrated as including memory/storage 1410 .
- the memory/storage 1410 represents memory/storage capacity associated with one or more computer-readable media.
- the memory/storage 1410 includes one or more of volatile media (such as random access memory (RAM)), nonvolatile media (such as read only memory (ROM), Flash memory, optical disks, magnetic disks.
- the memory/storage 1410 includes one or more of fixed media (e.g., RAM, ROM, a fixed hard drive), removable media (e.g., Flash memory, a removable hard drive, an optical disc).
- fixed media e.g., RAM, ROM, a fixed hard drive
- removable media e.g., Flash memory, a removable hard drive, an optical disc
- Input/output interface(s) 1406 are representative of functionality to allow a user to enter commands and information to computing device 1400 , and also allow information to be presented to the user and/or other components or devices using various input/output devices.
- Examples of input devices include a keyboard, a cursor control device (e.g., a mouse), a microphone (e.g., for voice recognition and/or spoken input), a scanner, touch functionality (e.g., capacitive or other sensors that are configured to detect physical touch), a camera (e.g., which may employ visible or non-visible wavelengths such as infrared frequencies to detect movement that does not involve touch as gestures), and so forth.
- a keyboard e.g., a cursor control device (e.g., a mouse), a microphone (e.g., for voice recognition and/or spoken input), a scanner, touch functionality (e.g., capacitive or other sensors that are configured to detect physical touch), a camera (e.g., which may employ visible or non-visible wavelengths such as infrared frequencies to detect movement that does not involve touch as gestures), and so forth.
- a cursor control device e.g., a mouse
- microphone e.g., for voice recognition and
- Examples of output devices include a display device (e.g., a monitor or projector), speakers, a printer, a network card, tactile-response device, and so forth.
- a display device e.g., a monitor or projector
- speakers e.g., a speaker
- printer e.g., a printer
- network card e.g., a network card
- modules include routines, programs, objects, elements, components, data structures, and so forth that perform particular tasks or implement particular abstract data types.
- module generally represent software, firmware, hardware, or a combination thereof.
- the features of the techniques described herein are platform-independent, meaning that the techniques may be implemented on a variety of commercial computing platforms having a variety of processors.
- an implementation of the described modules and techniques is stored on or transmitted across some form of computer-readable media.
- the computer-readable media includes a variety of media that may be accessed by the computing device 1400 .
- Hardware elements 1408 and computer-readable media 1404 are representative of instructions, modules, programmable device logic and/or fixed device logic implemented in a hardware form that may be employed in some embodiments to implement at least some aspects of the techniques described herein.
- hardware elements include one or more of components of an integrated circuit or on-chip system, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a complex programmable logic device (CPLD).
- ASIC application-specific integrated circuit
- FPGA field-programmable gate array
- CPLD complex programmable logic device
- a hardware element operates as a processing device that performs program tasks defined by instructions, modules, and/or logic embodied by the hardware element as well as a hardware device utilized to store instructions for execution, for example computer-readable storage media.
- software, hardware, or program modules and other program modules are implemented as one or more instructions and/or logic embodied on some form of computer-readable storage media and/or by one or more hardware elements 1408 .
- the computing device 1400 is configured to implement particular instructions and/or functions corresponding to the software and/or hardware modules. Accordingly, implementation of modules that are executable by the computing device 1400 as software may be achieved at least partially in hardware, e.g., through use of computer-readable storage media and/or hardware elements 1408 of the processing system.
- instructions and/or functions are executable/operable by one or more computing devices 1400 and/or processing systems 1402 to implement techniques, modules, and examples described herein.
- the computing device 1400 comprises a device selected from a computer class of devices that includes personal computers, desktop computers, multi-screen computers, laptop computers, netbooks. In an embodiment the computing device 1400 comprises a device selected from a mobile class of devices that includes mobile phones, tablet computers, multi-screen computers, wearable devices.
- the module 300 and related components is implemented at least partly through use of a distributed system over a cloud 1412 via a platform 1414 for resources 1416 .
- the platform 1414 abstracts underlying functionality of hardware, for example servers, and software resources of the cloud 1412 .
- resources 1416 include applications and/or data that are used while computer processing is executed on servers that are remote from the computing device 1400 .
- resources 1416 include services provided over the Internet and/or through a subscriber network, such as a cellular or Wi-Fi network.
- platform 1414 abstracts resources and functions to connect the computing device 1400 with other computing devices.
- the platform 1414 serves to abstract scaling of resources to provide a corresponding level of scale to encountered demand for the resources 1416 that are implemented via the platform 1414 .
- implementation of functionality described herein tends to be distributed.
- the functionality is implemented in part on the computing device 1400 as well as via the platform 1414 that abstracts the functionality of the cloud 1412 .
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Animal Behavior & Ethology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Pathology (AREA)
- Molecular Biology (AREA)
- Physics & Mathematics (AREA)
- Surgery (AREA)
- Biophysics (AREA)
- Physiology (AREA)
- Signal Processing (AREA)
- Psychiatry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Pulmonology (AREA)
- Anesthesiology (AREA)
- Hematology (AREA)
- Emergency Medicine (AREA)
- Power Engineering (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
A method and system of determining a sleep stage of a user involves receiving a respiratory flow signal of a user, obtaining at least one respiratory feature from at least part of the respiratory flow signal, and determining a sleep stage from the at least one respiratory feature.
Description
- This application is a continuation of U.S. patent application Ser. No. 15/373,344, filed Dec. 8, 2016, which claims priority to U.S. Provisional Application No. 62/264,787, filed Dec. 8, 2015, which is hereby incorporated by reference herein in its entirety.
- Any and all applications for which a foreign or domestic priority claim is identified in the Application Data Sheet as filed with the present application are hereby incorporated by reference under 37 CFR 1.57.
- The invention relates to a method and system for sleep stage determination. More particularly the invention relates to sleep stage determination based at least partly on respiratory signal.
- Devices or systems for providing a humidified gases flow to a user for therapeutic purposes are well known in the art. Such devices are typically configured to detect Sleep Disordered Breathing (SDB) events. Such devices are not usually configured to obtain more detailed information about a user's sleep patterns. One example is the determination of sleep stage of a user.
- Sleep stage is determined by the analysis of bio-signals including at least (electroencephalogram (eeg), electromyogram (emg), and electro-oculogram (eog). The definition of sleep stages is based on the characteristics of the bio-signals.
- At the present time, the most accepted method to determine sleep stages requires the interaction from experienced sleep technicians, which analyses several traces of signals (PSG study).
- It is an object of at least preferred embodiments of the present invention to address some of the aforementioned disadvantages. An additional or alternative object is to at least provide the public with a useful choice.
- In one aspect a method of determining a sleep stage of a user comprises receiving a respiratory flow signal of a user; obtaining at least one respiratory feature from at least part of the respiratory flow signal; and determining a sleep stage from the at least one respiratory feature.
- The term ‘comprising’ as used in this specification means ‘consisting at least in part of’. When interpreting each statement in this specification that includes the term ‘comprising’, features other than that or those prefaced by the term may also be present. Related terms such as ‘comprise’ and ‘comprises’ are to be interpreted in the same manner.
- Preferably the at least one respiratory feature is based at least partly on at least one duration measurement.
- Preferably the at least one duration measurement includes one or more of breath duration, inspiration duration, maximum inspiration time, maximum expiration time, a function of maximum expiration time and maximum inspiration time, a function of inspiration duration and breath duration.
- Preferably the at least one respiratory feature is based at least partly on at least one amplitude measurement.
- Preferably the at least one amplitude measurement includes one or more of maximum inspiration amplitude, maximum expiration amplitude, a function of maximum inspiration amplitude and maximum expiration amplitude.
- Preferably the at least one respiratory feature is based at least partly on at least one centre of mass related measurement.
- Preferably the at least one centre of mass related measurement includes one or more of inspiration centre of mass time, inspiration centre of mass amplitude, expiration centre of mass time, expiration centre of mass amplitude, a function of expiration centre of mass time and inspiration centre of mass time, a function of expiration centre of mass amplitude and inspiration centre of mass amplitude.
- Preferably the at least one respiratory feature is based at least partly on at least one derivative related measurement.
- Preferably the at least one derivative related measurement includes one or more of maximum negative acceleration time, maximum negative acceleration amplitude, maximum positive acceleration time, maximum positive acceleration amplitude, maximum negative flow rate time, maximum negative flow rate amplitude, maximum inspiration acceleration time, maximum inspiration acceleration amplitude.
- Preferably the at least one respiratory feature is based at least partly on at least one volume related measurement.
- Preferably the at least one volume related measurement includes one or more of inspiration volume, expiration volume, a function of inspiration volume and expiration volume.
- Preferably the method further comprises identifying, within the respiratory flow signal, at least one breath signal representing a breath of the user; and obtaining at least one breath measurement from a portion of the respiratory flow signal within which the at least one breath signal is identified.
- Preferably the method further comprises identifying, within the respiratory flow signal, a window containing a plurality of breath signals; and obtaining respective breath measurements of the breath signals within the window.
- Preferably the at least one breath feature comprises a mean and/or standard deviation of the breath measurements within at least part of the window.
- Preferably the method further comprises determining a sleep stage from the at least one breath feature at least partly by applying at least one of a supervised learning algorithm, an unsupervised learning algorithm, a semi-supervised learning algorithm.
- Preferably the at least one supervised learning algorithm comprises one or more of linear and logistic regression, support vector machine, artificial neural network, decision tree.
- Preferably the sleep stage comprises one of awake, N1, N2, N3, REM.
- Preferably the sleep stage comprises one of awake, light sleep, deep sleep, REM.
- Preferably the sleep stage comprises one of awake, non-REM, REM.
- Preferably the sleep stage comprises one of awake, sleep.
- Preferably the method further comprises applying a filter to the respiratory flow signal to remove at least one of high frequency noise, DC level.
- In another aspect, a sleep determination system comprises a feature extractor configured to obtain at least one respiratory feature from at least part of a respiratory flow signal of a user; and a mapping module configured to determine a sleep stage from the at least one respiratory feature.
- In another aspect, a sleep determination system comprises a processor; and a computer readable medium having stored thereon computer executable instructions that, when executed by the processor, cause the processor to perform a method of determining a sleep stage of a user. The method comprises receiving a respiratory flow signal of a user; obtaining at least one respiratory feature from at least part of the respiratory flow signal; and determining a sleep stage from the at least one respiratory feature.
- In another aspect, a computer readable medium has stored thereon computer-executable instructions that, when executed by a processor, cause the processor to perform a method of determining a sleep stage of a user. The method comprises receiving a respiratory flow signal of a user; obtaining at least one respiratory feature from at least part of the respiratory flow signal; and determining a sleep stage from the at least one respiratory feature.
- The invention in one aspect comprises several steps. The relation of one or more of such steps with respect to each of the others, the apparatus embodying features of construction, and combinations of elements and arrangement of parts that are adapted to affect such steps, are all exemplified in the following detailed disclosure.
- To those skilled in the art to which the invention relates, many changes in construction and widely differing embodiments and applications of the invention will suggest themselves without departing from the scope of the invention as defined in the appended claims. The disclosures and the descriptions herein are purely illustrative and are not intended to be in any sense limiting. Where specific integers are mentioned herein which have known equivalents in the art to which this invention relates, such known equivalents are deemed to be incorporated herein as if individually set forth.
- As used herein, ‘(s)’ following a noun means the plural and/or singular forms of the noun.
- As used herein, the term ‘and/or’ means ‘and’ or ‘or’ or both.
- It is intended that reference to a range of numbers disclosed herein (for example, 1 to 10) also incorporates reference to all rational numbers within that range (for example, 1, 1.1, 2, 3, 3.9, 4, 5, 6, 6.5, 7, 8, 9, and 10) and also any range of rational numbers within that range (for example, 2 to 8, 1.5 to 5.5, and 3.1 to 4.7) and, therefore, all sub-ranges of all ranges expressly disclosed herein are hereby expressly disclosed. These are only examples of what is specifically intended and all possible combinations of numerical values between the lowest value and the highest value enumerated are to be considered to be expressly stated in this application in a similar manner.
- In this specification where reference has been made to patent specifications, other external documents, or other sources of information, this is generally for the purpose of providing a context for discussing the features of the invention. Unless specifically stated otherwise, reference to such external documents or such sources of information is not to be construed as an admission that such documents or such sources of information, in any jurisdiction, are prior art or form part of the common general knowledge in the art.
- Although the present invention is broadly as defined above, those persons skilled in the art will appreciate that the invention is not limited thereto and that the invention also includes embodiments of which the following description gives examples.
- The term ‘connected to’ as used in this specification in relation to data or signal transfer includes all direct or indirect types of communication, including wired and wireless, via a cellular network, via a data bus, or any other computer structure. It is envisaged that they may be intervening elements between the connected integers. Variants such as ‘in communication with’, ‘joined to’, and ‘attached to’ are to be interpreted in a similar manner. Related terms such as ‘connecting’ and ‘in connection with’ are to be interpreted in the same manner.
- The term ‘computer-readable medium’ should be taken to include a single medium or multiple media. Examples of multiple media include a centralised or distributed database and/or associated caches. These multiple media store the one or more sets of computer executable instructions. The term ‘computer readable medium’ should also be taken to include any medium that is capable of storing, encoding or carrying a set of instructions for execution by a processor and that cause the processor to perform any one or more of the methods described above. The computer-readable medium is also capable of storing, encoding or carrying data structures used by or associated with these sets of instructions. The term ‘computer-readable medium’ includes solid-state memories, optical media and magnetic media.
- Preferred forms of the method and system for sleep stage determination will now be described by way of example only with reference to the accompanying figures in which:
-
FIG. 1 is a perspective view of a Continuous Positive Airway Pressure (CPAP) machine configured to obtain a respiratory flow signal from a user; -
FIG. 2 shows a schematic view of the CPAP machine ofFIG. 1 ; -
FIG. 3 shows a schematic view of a sleep stage determination module; -
FIG. 4 shows an example of a method performed by the sleep stage determination module ofFIG. 3 ; -
FIG. 5 shows examples of duration measurements and amplitude measurements from which the feature extractor ofFIG. 3 obtains respiratory feature(s). -
FIG. 6 shows examples of centre of mass measurements from which the feature extractor obtains respiratory feature(s). -
FIG. 7 shows examples of derivative related measurements from which the feature extractor obtains respiratory feature(s). -
FIG. 8 shows examples of volume related measurements from which the feature extractor obtains respiratory feature(s). -
FIG. 9 shows an example of the feature extractor obtaining respiratory feature(s) from window(s) comprising multiple breath signals. -
FIG. 10 shows an example of agreement between the mapping module ofFIG. 3 and the inferred sleep stages of a test study. -
FIG. 11 shows an example function of the context module fromFIG. 3 applying contextual clues in an attempt to improve accuracy of the performance of the mapping module. -
FIG. 12 shows the accuracy per patient of a test study. -
FIG. 13 shows an example of typical performance for an individual patient. -
FIG. 14 shows an example of a computing system and/or device that implements the sleep stage determination module ofFIG. 3 . -
FIG. 1 shows auser 100 receiving air from a modular assisted breathing unit andhumidifier system 102.System 102 is shown inFIG. 1 as a Continuous Positive Airway Pressure (CPAP) machine. In an embodiment thesystem 102 provides a pressurised stream of heated, humidified gases to theuser 100 for therapeutic purposes. These therapeutic purposes include for example one or more of reducing the incidence of obstructive sleep apnea, providing CPAP therapy, providing humidification for therapeutic purposes. - A heated and humidified gases stream passes along the length of a
delivery conduit 104 and are provided to theuser 100 via auser interface 106. In an embodiment theconduit 104 is heated via a heater wire (not shown) or similar to help prevent rain-out. - The
conduit 104 typically has a circular internal cross-section. The internal diameter of the conduit is typically about 20 mm. in an embodiment the internal diameter is between 10 mm and 30 mm. These typical dimensions apply to both flexible portions of the gases flow passageway and rigid components such as elbows and connectors and portions integrated into components of the humidified gases supply. - In an embodiment the
user interface 106 comprises a full face mask. It will be appreciated that alternatives to a full face mask include a nasal mask that surrounds and covers the nose of theuser 100, nasal cannula, tracheostomy fitting, any other suitable user interface. -
FIG. 2 shows a schematic view ofsystem 100 ofFIG. 1 . Thesystem 100 includes anair supply 200 configured to deliver the heated and humidified gases stream into thedelivery conduit 104. - In an embodiment the
air supply 200 includes a flow control mechanism that in turn includes for example an internal compressor unit and a flow generator orfan unit 202. Air from the atmosphere enters the housing of theair supply 200 via an atmospheric inlet (not shown) and is drawn through thefan unit 202. In an embodiment the output of thefan unit 202 is adjustable by varying the speed of the fan. - In an embodiment the
air supply 200 includes or is interfaced to a humidifier chamber (not shown). The humidifier chamber contains a volume of water. A heater plate heats the base of the humidifier chamber thereby heating the contents of the chamber. As the water in the chamber is heated it evaporates and the gases within the humidifier chamber become heated and humidified. The gases stream entering the humidifier chamber passes over the heated water and becomes heated and humidified as it does so. - As described above the heated and humidified gases stream enters the
delivery conduit 104. - A
central controller 204 or control system is connected to amemory 206. Thecontroller 204 receives user input signals viauser input 208 and displays output to a user viadisplay 210. In an embodiment thecontroller 204 receives input from at least onesensor 212 located at various points withinsystem 102,conduit 104 and/orinterface 106. In response to the input fromuser input 208 and sensor(s) 212 thecontroller 204 determines a control output that sends signals to adjust the power to theair supply 202. - In an embodiment the sensor(s) 212 receive signals representative of inspiration and expiration of the
user 100. The sensor(s) 212 sends a respiratory flow signal to thecontroller 204. -
FIG. 3 shows a schematic view of a sleepstage determination module 300. Themodule 300 receives arespiratory flow signal 302. An example of a respiratory flow signal is one sent from the sensor(s) 212 to thecontroller 204. - In an embodiment a
filter 304 pre-processes the respiratory flow signal to remove high frequency noise and DC level. The use of thefilter 304 is optional. In an embodiment thefilter 304 is not present or therespiratory flow signal 302 is not sent to thefilter 304. - In an embodiment a
flow segmenter 306 segments the respiratory flow signal into time-based epochs and/or breath-number-based epochs for subsequent feature extraction as will be further described below. The use of theflow segmenter 306 is optional. In an embodiment feature extraction is performed on the rawrespiratory flow signal 302. - In an embodiment a flow
segment measurement module 308 determines measurements of therespiratory flow signal 302 on which feature extraction is subsequently based as will be described below. In an embodiment themeasurement module 308 determines one or more of duration measurements, amplitude measurements, centre of mass related measurements, derivative related measurements, volume related measurements. - In an embodiment a
feature extractor 310 obtains at least one respiratory feature from therespiratory flow signal 302. As described above the feature extractor takes as input one or more of a filtered respiratory flow signal, a segmented respiratory flow signal in which breath segmentation has been performed, an otherwise unprocessed respiratory flow signal. - In an embodiment a
mapping module 312 maps a set of respiratory features to at least one sleep stage as will be further described below. Acontext module 314 optionally applies context data in an attempt to improve the accuracy of themapping module 312 output. - The
module 300 outputs asleep stage 316 determined by themapping module 312 andoptionally context module 314 from therespiratory flow signal 302. In an embodiment the module determines a sequence of and/or multiple sleep stages 316 from arespiratory flow signal 302. - In an embodiment the
module 300 is implemented within the modular assisted breathing unit andhumidifier system 102. The module outputs thesleep stage 316 on thedisplay 210 and/or stores the sleep stage on thememory 206. - In an embodiment the
module 300 is implemented on a computing device for example thecomputing device 1400 ofFIG. 14 . Thesleep stage 316 is presented on a display and/or maintained on a memory that either form(s) part of thecomputing device 1400 or is/are connected tocomputing device 1400. -
FIG. 4 shows an example of amethod 400 performed by the sleepstage determination module 300 ofFIG. 3 . In an embodiment, themodule 300 receives 402 a respiratory flow signal. Filtering (not shown) is optionally performed on the respiratory flow signal to remove high frequency noise and DC level. - In an embodiment the
flow segmenter 306 ofFIG. 3 optionally applies 404 flow segmentation to the respiratory flow signal. As described above, the flow segmenter segments the respiratory flow signal into time-based epochs and/or breath-number-based epochs for subsequent feature extraction. - In an embodiment the
measurement module 308 obtains 406 flow segment measurements of the respiratory flow signal on which feature extraction is subsequently based. Examples of measurements include one or more of duration measurements, amplitude measurements, centre of mass related measurements, derivative related measurements, volume related measurements. In an embodiment these measurements comprise breath measurements. - In an embodiment the
feature extractor 310 performs 408 feature extraction on flow segment measurements. For example, thefeature extractor 310 obtains at least one respiratory feature from the respiratory flow signal. - The
module 300 determines 410 a sleep stage, a sequence of sleep stages, and/or multiple sleep stages from the features extracted from the respiratory flow signal. In an embodiment this function is performed by themapping module 312 and optionally also thecontext module 314. - In an embodiment, the feature extractor obtains at least one respiratory feature based at least partly on at least one duration measurement.
FIG. 5 shows examples of duration measurements associated to amaximum inspiration 500 and amaximum expiration 502. - In an embodiment, the duration measurements include one or more of:
-
- breath duration
- inspiration duration
- maximum inspiration time
- maximum expiration time
- In an embodiment, the duration measurements include a function of two or more duration measurements. Examples of functions include:
-
- a difference between maximum expiration time and maximum inspiration time
- a ratio of inspiration duration to breath duration.
- In an embodiment, the feature extractor obtains at least one respiratory feature based at least partly on at least one amplitude measurement.
FIG. 5 also shows examples of amplitude measurements. - In an embodiment, the amplitude measurements include one or more of:
-
- maximum inspiration amplitude
- maximum expiration amplitude.
- In an embodiment, the amplitude measurements include a function of two or more amplitude measurements. An example of a function is a difference between maximum inspiration amplitude and maximum expiration amplitude.
- In an embodiment the duration and/or amplitude measurements are obtained from a single breath. In an embodiment the duration and/or amplitude measurements are calculated as a mean or average of at least two breaths. In an embodiment the at least two breaths are consecutive.
- In an embodiment, the feature extractor obtains at least one respiratory feature based at least partly on at least one centre of mass measurement.
FIG. 6 shows examples of centre of mass measurements. - In an embodiment, the centre of mass measurement includes inspiration centre of mass time. The centre of mass of the body formed by the inspiration curve is indicated at 600.
- A time axis component of the inspiration centre of
mass 600 is calculated for example relative to the beginning of the breath by the following: -
- where NInsp is the number of inspiration points i in the curve, yi and xi are the signal amplitudes and time, respectively.
- In an embodiment, the centre of mass measurement includes expiration centre of mass time. The centre of mass of the body formed by the inspiration curve is indicated at 602. A time axis component of the expiration centre of
mass 602 is calculated for example relative to the beginning of the breath by the following: -
- where NExp is the number of expiration points i in the curve, yi and xi are the signal amplitude and time respectively.
- In an embodiment, the centre of mass measurement includes a function of expiration centre of mass time and inspiration centre of mass time. An example is the difference between expiration centre of mass time and inspiration centre of mass time.
- In an embodiment, the centre of mass measurement includes inspiration centre of mass amplitude. An amplitude axis component of the inspiration centre of
mass 600 is calculated for example relative to the inspiration curve and the baseline by the following: -
- where NInsp is the number of inspiration points i in the curve, yi and xi are the signal amplitude and time, respectively.
- In an embodiment, the centre of mass measurement includes expiration centre of mass amplitude. An amplitude axis component of the expiration centre of
mass 602 is calculated for example relative to the expiration curve and the baseline by the following: -
- where NExp is the number of expiration points i in the curve, yi and xi are the signal amplitude and time, respectively.
- In an embodiment, the centre of mass measurement includes a function of expiration centre of mass amplitude and inspiration centre of mass amplitude. An example is the difference between expiration centre of mass amplitude and inspiration centre of mass amplitude.
- In an embodiment, the feature extractor obtains at least one respiratory feature based at least partly on at least one derivative related measurement.
FIG. 7 shows examples of derivative related measurements. - In an embodiment, the derivative related measurement includes a maximum negative acceleration time.
FIG. 7 shows amaximum inspiration 700 and amaximum expiration 702. A maximumnegative acceleration 704 has a time value between the time ofmaximum inspiration 700 and the time ofmaximum expiration 702. The time value of 704 is a time when the acceleration reaches a negative peak, calculated from the beginning of the breath. In an embodiment the time value is computed with the second derivative obtained from a Savitzky Golay filter. - In an embodiment, the derivative related measurement includes a maximum positive acceleration time, having a time value between the time of
maximum inspiration 700 and the time ofmaximum expiration 702. The time value of 706 is a time when the acceleration reaches a positive peak, calculated from the beginning of the breath. In an embodiment the time value of 706 is computed with the second derivative obtained from a Savitzky Golay filter. - In an embodiment, the derivative related measurement includes a time value for a
maximum inspiration acceleration 710. A time value formaximum inspiration acceleration 710 is a time value between the start of the breath andmaximum inspiration 700 when the acceleration is at a maximum. In an embodiment the time value is computed with the second derivative obtained from a Savitzky Golay filter. - In an embodiment the derivative related measurement includes a function of a maximum negative acceleration time and a maximum positive acceleration time. An example is a maximum negative
flow rate time 708 comprising a time between maximum negative acceleration and maximum positive acceleration when the acceleration is zero, calculated from the beginning of the breath. In an embodiment this is computed by finding a first zero crossing position on a second derivative signal generated by a Savitsky Golay filter. - In an embodiment, the derivative related measurement includes a maximum negative acceleration amplitude. The amplitude of the maximum
negative acceleration 704 occurs between a time ofmaximum inspiration 700 and a time ofmaximum expiration 702. In an embodiment the amplitude of maximumnegative acceleration 704 is computed with the second derivative obtained from a Savitzky Golay filter. - In an embodiment, the derivative related measurement includes a maximum positive acceleration amplitude. The amplitude of the maximum
positive acceleration 706 occurs between a time of maximumnegative acceleration 704 and the end of the breath. In an embodiment the amplitude of maximumpositive acceleration 706 is computed with the second derivative obtained from a Savitzky Golay filter. - In an embodiment, the derivative related measurement includes an amplitude for the
maximum inspiration acceleration 710. The amplitude occurs between the start of the breath andmaximum inspiration 700. In an embodiment the amplitude is computed with the second derivative obtained from a Savitzky Golay filter. - In an embodiment the derivative related measurement includes a function of a maximum negative acceleration time and a maximum positive acceleration time. An example is an amplitude of maximum
negative flow rate 708 comprising an amplitude of a maximum rate of change in the flow signal that occurs between maximum negative acceleration and maximum positive acceleration. In an embodiment the amplitude is approximated by calculating the difference in flow at the point when the acceleration is zero. - In an embodiment, the feature extractor obtains at least one respiratory feature based at least partly on at least one volume related measurement.
FIG. 8 shows examples of volume related measurements. - In an embodiment, the volume related measurement includes
inspiration volume 800 comprising the volume of the inspiration part of the flow signal above the baseline. Theinspiration volume 800 is calculated for example by the following: -
- where NInsp is the number of inspiration points i in the curve, yi and B are the signal amplitude and the baseline, respectively. SR is the sampling rate of the signal in samples per minute.
- In an embodiment, the volume related measurement includes
expiration volume 802 comprising the volume of the inspiration part of the flow signal above the baseline. Theexpiration volume 802 is calculated for example by the following: -
- where NExp is the number of expiration points i in the curve, yi and B are the signal amplitude and the baseline, respectively. SR is the sampling rate of the signal in samples per minute.
- In an embodiment, the volume related measurement includes a function of inspiration volume and expiration volume. An example is a ratio between inspiration volume and expiration volume.
-
FIG. 9 shows an example of feature extraction that in an embodiment is performed by thefeature extractor 310 ofFIG. 3 . Awindow 900 comprises a plurality of breath signals within a respiratory flow signal.Window 900 is shown centred on breath B indicated at 902. - In an embodiment the window is of size N so that it includes N consecutive breaths immediately following breath B in the respiratory flow signal, and includes N consecutive breaths immediately preceding breath B in the respiratory flow signal. The boundaries of
window 900 are defined so as to include breath+N indicated at 904, and breath−N indicated at 906. - In an embodiment the value of N is 15. In an embodiment the value of N is selected from a range of 2 to 30.
- In an embodiment the
window 900 is further subdivided into at least one further window, forexample window 908 andwindow 910. In anembodiment window 908 is ⅔ the size ofwindow 900. In anembodiment window 910 is ⅓ the size ofwindow 900. - In an embodiment,
window 908 and/orwindow 910 has/have boundaries defined so as to include breath B and breath+N but not breath−N. In an embodiment,window 908 and/orwindow 910 has/have boundaries defined so as to include breath B and breath−N but not breath+N. In an embodiment,window 908 and/orwindow 910 has/have boundaries defined so as to include breath B but not breath+N and not breath−N. - One example of a respiratory feature obtained from at least part of the respiratory flow signal is a mean breath measurement associated to at least some of the breaths in the respiratory flow signal.
- In an embodiment the respiratory features obtained from the respiratory flow signal represent a mean breath measurement associated to the breaths within one or more of
window 900,window 908,window 910. - Another example of a respiratory feature obtained from at least part of the respiratory flow signal is a standard deviation of breath measurements associated to at least some of the breaths in the respiratory flow signal.
- In an embodiment the respiratory features obtained from the respiratory flow signal represent a standard deviation of breath measurements associated to the breaths within one or more of
window 900,window 908,window 910. - In an embodiment the breath measurements include one or more of duration measurements, amplitude measurements, centre of mass related measurements, derivative related measurements, volume related measurements.
- In an embodiment the breath features describe the morphology of part of the respiratory flow signal that surrounds a breath B under evaluation. In an embodiment the mean and/or standard deviation for up to 26 breath measurements are determined for one or more of
windows - In an embodiment, a mapping is performed by the
mapping module 312 ofFIG. 3 . Themapping module 312 is configured to map a given set of respiratory features to a sleep stage. - In an embodiment the
mapping module 312 applies a supervised learning algorithm. Suitable algorithms include one or more of linear and logistic regression, support vector machine, artificial neural network, decision tree. In an embodiment, the learning is performed offline with previously recorded and labelled data. - In an embodiment the
mapping module 312 applies an unsupervised learning algorithm. In an embodiment themapping module 312 applies a semi-supervised learning algorithm. - In an embodiment, machine learning methods that provide such learning capabilities are used to improve accuracy of the
module 300. In an embodiment, learning is carried out to suit the conditions of individual patients. For instance, parts of the signal where strong markers are present, for example start of the night or periods prior to events, can be used to adapt the model to specific patient profiles. - In an embodiment the
mapping module 312 maps a set of respiratory features to a sleep stage selected from awake, N1, N2, N3, REM. In an embodiment the sleep stage is selected from awake, light sleep, deep sleep, REM. In an embodiment the sleep stage is selected from awake, non-REM, REM. In an embodiment the sleep stage is selected from awake, sleep. - In an embodiment the features are calculated on a breath basis. In this case the
module 300 provides breath-basis output. The breath outputs are then combined in 30 second epochs to calculate agreement to a correct hypnogram determined bymapping module 312. The resultant epoch output for each class is calculated as the average of breaths output. The resultant epoch output is then normalized so that the sum of the outputs for all classes are 1. - In an embodiment a
context module 314 is configured to apply contextual clues in an attempt to improve accuracy of themapping module 312. - One example of a contextual clue is identification of periods without breathing. It can be assumed that a user is awake after long periods in which the absence of a respiratory flow signal is detected. This occurs in the beginning of the session and when the user takes the mask off during the night. An example threshold is 7 minutes. After this threshold period of no respiratory flow signal it is assumed the user is awake and a sleep stage of awake is determined.
- Another example of a contextual clue is identification of Sleep Disordered Breathing (SDB) events. It is assumed that prior to the detection of SDB events, the user is asleep. Prior to SDB events, a sleep stage of sleep is set. This could be light, deep, or REM. In an embodiment a threshold window of 10 epochs (5 minutes) prior to an SDB event is used.
- Another example of a contextual clue is neighbouring consensus. It is assumed that individual epochs with differing labels compared to their neighbouring epochs are wrongly detected. One aim is to remove sleep stages that are likely to have been wrongly detected, by analysing the context where they are placed.
- In an embodiment, the modelling of sleep stages is performed using algorithms that learn temporal structures at different scales. Hidden Markov Models (HMM) and its state machine representation and Temporal Delay Neural Networks (TDNN) are examples of learning methods that can be used. In an embodiment, HMM is applied in an attempt to learn the sleep stage transition probabilities using several bio-signals. In an embodiment, ‘a priori’ probability is implemented to capture probabilities of observing a particular state at different periods of the night.
- In an embodiment, the
module 300 implements a sleep score. The sleep score attempts to capture, in at least one index, sleep quality and effectiveness. In an embodiment, multiple indices are used. - There is no clear consensus about the best metrics to capture quality of sleep. One reason for this is that the requirements for having a good night of sleep can vary from person to person, due to age, external factors, etc. Some metrics have been proposed, each one capturing the sleep condition in different ways. Some of these metrics are:
-
- Sleep efficiency—Sleep efficiency is defined as Total Sleep time/Total time in bed. It captures the amount of time effectively sleeping while in bed.
- Deep sleep time—Periods of deep sleep are when the body relaxes and are usually associated with body and muscular recovery (physical restoration).
- REM time—Period of REM time are usually associated memory and learning consolidation.
- Sleep fragmentation—Can be defined in different ways, but essentially tries to capture the number of times arousals/awakes occur during the night. Highly fragmented sleep is usually associated with poor sleep and difficulty to remain in deep sleep or REM stage for long periods.
- A good night of sleep should ideally have long periods of deep sleep and REM stages and low sleep fragmentation.
- In an embodiment the
module 300, or a further module taking input frommodule 300, calculates one or more of the indices Sleep time (ST), Total Time at Deep Sleep (DST), Total Time at REM (REMT), Number of Sleep Disruptions (number of awakes during the night after sleep onset) (NSD), Number of SDB events (NSDB). - In an embodiment the indices are combined as:
- where:
- ST score changes linearly [0 for 0 hours, 100 for 7 hours or more]
- DST score changes linearly [0 for 0 hours, 100 for 1.5 hours or more]
- REMT score changes linearly [0 for 0 hours, 100 for 1.5 hours or more)
- NSD score changes linearly [100 for 0, 0 for 10 or more disruptions]
- NSDB score changes linearly [100 for 0, 0 for 10 or more events/hour].
- Setting X1 to X5 to 0.2, equal weight is given to each score and the resultant sleep score is in the range [0, 1].
- Experimental results are described with reference to
FIGS. 10-13 . Described below is an evaluation of the performance of a system to determine partial sleep staging from the respiratory flow signal alone. In particular, described is the performance of models automatically generated based on flow patterns. - A dataset composed of 50 studies, with full PSG recordings, is used in the experiments. All patients in the study use a modular assisted breathing unit and
humidifier system 102 at fixed pressure or pressure auto adjustment mode. - In total, more than 370000 breaths were detected using a breath detector algorithm. To speed up the process of training and testing, half of the samples were used (every second sample). From the remaining 187806 samples, at most 400 samples from each patient were used for training ( 100 samples from each class), resulting in a training pool of 19607 breaths. The number of training samples is lower than 50×400=20000 because 100 samples of each class were not available in all patients. The test pool was composed of the 168199 samples.
- Five-folds cross-validation technique was used for error estimation, where the folds were stratified by patient, i.e., each fold is composed of samples from 10 patients.
- During the training procedure, a cost learning matrix was used to give different costs for each error as:
-
Awake Light Sleep Deep Sleep REM Awake 0 5 5 10 Light Sleep 5 0 5 5 Deep Sleep 5 5 0 5 REM 10 5 5 0 - The use of the cost learning matrix aims at reducing errors between specific classes. The aim is to direct the learning procedure to penalise the misclassification between Awake and REM.
- Once the training is done and the model is inferred, the operating point is chosen targeting equal TP Rate for all classes. The results of the cross-validated test sets have been used to find the equal TP Rate operating points. Having equal TP Rate in all classes implies that similar numbers of samples are allocated to each class when compared to the results of the
mapping module 312. - The best result was obtained with the Random Forest algorithm. In the
algorithm 100 decision trees were created and the final classification obtained by combining the results of each tree. The following table shows the resultant confusion matrix. The overall Accuracy was 0.559. -
Confusion matrix TP Rate Precision 3227 2192 288 172 Awake 0.549 0.599 1749 13620 6170 2288 Light Sleep 0.571 0.643 223 3381 4707 657 Deep Sleep 0.524 0.402 188 1964 541 3574 REM 0.570 0.534 -
FIG. 10 shows an example of agreement between themapping module 312 and the inferred sleep stages on a test study. -
FIG. 11 shows an application of thecontext module 314 to improve the accuracy of themapping module 312. - It can be assumed that individual epochs with differing labels compared to their neighbouring epochs are wrongly detected. This aims at removing states that are likely to have been wrongly detected, by analysing the context where they are placed.
- Using majority voting over neighbouring windows, for example recursively until no more changes occur, has the potential to replace extremely short-term stages with neighbouring labels. The number of neighbouring epochs N used for majority voting was set to 7.
- By using contextual cues the resultant accuracy increases to 0.602, shown as the confusion matrix in the following table.
-
Confusion matrix TP Rate Precision F- measure 3135 2383 197 124 Awake 0.537 0.663 0.593 1245 15180 5847 1548 Light Sleep 0.637 0.658 0.648 207 3434 4968 359 Deep Sleep 0.554 0.437 0.489 145 2067 321 3731 REM 0.596 0.648 0.620 -
FIG. 12 shows the accuracy per patient as having a range varying between [0.31, 0.83], mean=0.6 with 95% confidence that the mean is between [0.57, 0.63]. -
FIG. 13 shows an example of typical performance for an individual patient with agreement of 61.4%. -
FIG. 14 shows an example of a system representative of a computing system and/or device that implementsmodule 300 fromFIG. 3 . Thecomputing device 1400 comprises one or more of a server of a service provider, a device associated with the client (for example a client device), an on-chip system, any other suitable computing device or computing system. - In an embodiment,
computing device 1400 includes aprocessing system 1402, one or more computer-readable media 1404, and one or more Input/Output (I/O) Interfaces 1406 that are communicatively coupled, one to another. In an embodiment thecomputing device 1400 further includes a system bus or other data and command transfer system (not shown) that couples the various components, one to another. A system bus includes one or more of a memory bus or memory controller, a peripheral bus, a universal serial bus, a processor or local bus that utilizes any of a variety of bus architectures. - The
processing system 1402 is representative of functionality to perform one or more operations using hardware. Accordingly, theprocessing system 1402 is illustrated as including hardware element(s) 1408 configured as one or more of processors and/or functional blocks. This may include implementation in hardware as an application specific integrated circuit or other logic device formed using one or more semiconductors. Thehardware elements 1408 are not limited by the materials from which they are formed or the processing mechanisms employed therein. For example, processors may be comprised of semiconductor(s) and/or transistors (e.g., electronic integrated circuits (ICs)). In such a context, processor-executable instructions may be electronically-executable instructions - The computer-
readable media 1404 is illustrated as including memory/storage 1410. The memory/storage 1410 represents memory/storage capacity associated with one or more computer-readable media. In an embodiment the memory/storage 1410 includes one or more of volatile media (such as random access memory (RAM)), nonvolatile media (such as read only memory (ROM), Flash memory, optical disks, magnetic disks. - In an embodiment the memory/
storage 1410 includes one or more of fixed media (e.g., RAM, ROM, a fixed hard drive), removable media (e.g., Flash memory, a removable hard drive, an optical disc). - Input/output interface(s) 1406 are representative of functionality to allow a user to enter commands and information to
computing device 1400, and also allow information to be presented to the user and/or other components or devices using various input/output devices. - Examples of input devices include a keyboard, a cursor control device (e.g., a mouse), a microphone (e.g., for voice recognition and/or spoken input), a scanner, touch functionality (e.g., capacitive or other sensors that are configured to detect physical touch), a camera (e.g., which may employ visible or non-visible wavelengths such as infrared frequencies to detect movement that does not involve touch as gestures), and so forth.
- Examples of output devices include a display device (e.g., a monitor or projector), speakers, a printer, a network card, tactile-response device, and so forth.
- Various techniques are described herein in the general context of software, hardware elements, or program modules. Generally, such modules include routines, programs, objects, elements, components, data structures, and so forth that perform particular tasks or implement particular abstract data types. The terms ‘module’, ‘functionality’, and ‘component’ as used herein generally represent software, firmware, hardware, or a combination thereof. The features of the techniques described herein are platform-independent, meaning that the techniques may be implemented on a variety of commercial computing platforms having a variety of processors.
- In an embodiment, an implementation of the described modules and techniques is stored on or transmitted across some form of computer-readable media. The computer-readable media includes a variety of media that may be accessed by the
computing device 1400. -
Hardware elements 1408 and computer-readable media 1404 are representative of instructions, modules, programmable device logic and/or fixed device logic implemented in a hardware form that may be employed in some embodiments to implement at least some aspects of the techniques described herein. - In an embodiment, hardware elements include one or more of components of an integrated circuit or on-chip system, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a complex programmable logic device (CPLD). In an embodiment, a hardware element operates as a processing device that performs program tasks defined by instructions, modules, and/or logic embodied by the hardware element as well as a hardware device utilized to store instructions for execution, for example computer-readable storage media.
- In an embodiment, software, hardware, or program modules and other program modules are implemented as one or more instructions and/or logic embodied on some form of computer-readable storage media and/or by one or
more hardware elements 1408. Thecomputing device 1400 is configured to implement particular instructions and/or functions corresponding to the software and/or hardware modules. Accordingly, implementation of modules that are executable by thecomputing device 1400 as software may be achieved at least partially in hardware, e.g., through use of computer-readable storage media and/orhardware elements 1408 of the processing system. - In an embodiment the instructions and/or functions are executable/operable by one or
more computing devices 1400 and/orprocessing systems 1402 to implement techniques, modules, and examples described herein. - In an embodiment the
computing device 1400 comprises a device selected from a computer class of devices that includes personal computers, desktop computers, multi-screen computers, laptop computers, netbooks. In an embodiment thecomputing device 1400 comprises a device selected from a mobile class of devices that includes mobile phones, tablet computers, multi-screen computers, wearable devices. - In an embodiment the
module 300 and related components is implemented at least partly through use of a distributed system over acloud 1412 via aplatform 1414 forresources 1416. Theplatform 1414 abstracts underlying functionality of hardware, for example servers, and software resources of thecloud 1412. - In an embodiment the
resources 1416 include applications and/or data that are used while computer processing is executed on servers that are remote from thecomputing device 1400. In an embodiment,resources 1416 include services provided over the Internet and/or through a subscriber network, such as a cellular or Wi-Fi network. - In an embodiment,
platform 1414 abstracts resources and functions to connect thecomputing device 1400 with other computing devices. In an embodiment theplatform 1414 serves to abstract scaling of resources to provide a corresponding level of scale to encountered demand for theresources 1416 that are implemented via theplatform 1414. - In an interconnected device embodiment, implementation of functionality described herein tends to be distributed. For example, in an embodiment, the functionality is implemented in part on the
computing device 1400 as well as via theplatform 1414 that abstracts the functionality of thecloud 1412. - The foregoing description of the invention includes preferred forms thereof. Modifications may be made thereto without departing from the scope of the invention, as defined by the accompanying claims. 1.-19. (canceled)
Claims (20)
- 20. A method of automatically determining sleep quality and effectiveness of a user using a respiratory system configured to provide respiratory therapy to the user, the method comprising:using a controller of the respiratory system and one or more sensors of the respiratory system:receiving a respiratory flow signal of the user from the one or more sensors;obtaining one or more measurements from the respiratory flow signal;automatically determining a sleep stage of the user from the one or more measurements; andcalculating a sleep score based at least in part on the determined sleep stage of the user.
- 21. The method of
claim 20 , wherein the sleep score is calculated further based on one or more metrics: sleep efficiency, deep sleep time, rapid eye movement (REM) time, or sleep fragmentation. - 22. The method of
claim 21 , wherein the sleep efficiency is defined as total sleep time over total time in bed. - 23. The method of
claim 21 , wherein the sleep fragmentation is defined as a number of times arousals or awakes of the user occur during the night. - 24. The method of
claim 20 , wherein the sleep score is calculated further based on one or more indices: sleep time (ST), total time at deep sleep (DST), total time at REM (REMT), number of sleep disruptions (NSD), or number of sleep disordered breathing (SDB) events (NSDB). - 25. The method of
claim 24 , wherein the sleep score is calculated using an equation X1*ST+X2*DST+X3*REMT+X4*NSD+X5*NSDB, where X1-X5 are coefficients. - 26. The method of
claim 25 , where X1-X5 are each set to 0.2. - 27. The method of
claim 25 , where the sleep score is between 0 to 1. - 28. The method of
claim 20 , wherein the one or more measurements comprise at least one centre of mass measurement, at least one duration measurement, at least one amplitude measurement, or at least one volume related measurement. - 29. The method of
claim 20 , further comprising:identifying, within the respiratory flow signal, at least one breath signal representing a breath of the user; andobtaining at least one breath measurement from a portion of the respiratory flow signal within which the at least one breath signal is identified. - 30. The method of
claim 20 , further comprising:identifying, within the respiratory flow signal, a window containing a plurality of breath signals; andobtaining respective breath measurements of the plurality of breath signals within the window. - 31. The method of
claim 30 , wherein automatically determining the sleep stage of the user further comprises automatically determining the sleep stage of the user from at least one of a mean or a standard deviation of the breath measurements within at least part of the window. - 32. The method of
claim 20 , further comprising determining the sleep stage at least partly by applying at least one of a supervised learning algorithm, an unsupervised learning algorithm, or a semi-supervised learning algorithm. - 33. The method of
claim 20 , wherein the sleep stage is one of a plurality of sleep stages configured to be determined by the method, the plurality of sleep stages comprising:awake, N1, N2, N3, and REM;awake, light sleep, deep sleep, and REM;awake, non-REM, and REM; orawake and asleep. - 34. The method of
claim 20 , further comprising implementing a context module to apply contextual clues to improve accuracy of determining the sleep quality and effectiveness of the user. - 35. The method of
claim 34 , wherein the contextual clues comprise prolonged periods without breathing, the method further comprising determining the sleep stage is awake in response to detecting a prolonged period without breathing. - 36. The method of
claim 34 , wherein the contextual clues comprise SDB events, the method further comprising assuming the user is asleep prior to a detected SDB event and determining the sleep stage is light sleep, deep sleep, or REM in response to the detected SDB event. - 37. The method of
claim 34 , wherein the contextual clues comprise neighboring consensus, the method further comprising ignoring individual short periods of the respiratory flow signal that differ from neighboring portions of the respiratory flow signal. - 38. The method of
claim 20 , further comprising using one or more machine learning methods to identify markers in the respiratory flow signal for determining the sleep stage of the user. - 39. The method of
claim 38 , wherein the machine learning methods are configured to adapt determining of the sleep stage to specific user profiles.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US17/653,255 US20220313155A1 (en) | 2015-12-08 | 2022-03-02 | Flow-based sleep stage determination |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US201562264787P | 2015-12-08 | 2015-12-08 | |
US15/373,344 US11298074B2 (en) | 2015-12-08 | 2016-12-08 | Flow-based sleep stage determination |
US17/653,255 US20220313155A1 (en) | 2015-12-08 | 2022-03-02 | Flow-based sleep stage determination |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/373,344 Continuation US11298074B2 (en) | 2015-12-08 | 2016-12-08 | Flow-based sleep stage determination |
Publications (1)
Publication Number | Publication Date |
---|---|
US20220313155A1 true US20220313155A1 (en) | 2022-10-06 |
Family
ID=59275297
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/373,344 Active 2039-05-11 US11298074B2 (en) | 2015-12-08 | 2016-12-08 | Flow-based sleep stage determination |
US17/653,255 Pending US20220313155A1 (en) | 2015-12-08 | 2022-03-02 | Flow-based sleep stage determination |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/373,344 Active 2039-05-11 US11298074B2 (en) | 2015-12-08 | 2016-12-08 | Flow-based sleep stage determination |
Country Status (1)
Country | Link |
---|---|
US (2) | US11298074B2 (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20210113797A1 (en) * | 2017-12-20 | 2021-04-22 | Koninklijke Philips N.V. | Control unit for use with a respiratory assist device |
JP7407806B2 (en) * | 2018-10-01 | 2024-01-04 | コーニンクレッカ フィリップス エヌ ヴェ | Systems and methods for using respiratory events in sleep stages |
US11508484B1 (en) * | 2018-12-28 | 2022-11-22 | ResMed Pty Ltd | Prediction of respiratory therapy compliance |
US20210085242A1 (en) * | 2019-09-20 | 2021-03-25 | Nox Medical Ehf | System and method for determining sleep stages based on non-cardiac body signals |
CN111436904B (en) * | 2020-03-17 | 2022-11-15 | 深圳赛安特技术服务有限公司 | Intelligent sleep staging method and device and computer readable storage medium |
IT202000011221A1 (en) * | 2020-05-15 | 2021-11-15 | St Microelectronics Srl | SYSTEM AND METHOD OF DETECTING THE LIFTING AND LOWERING OF A USER'S FOOT FOR THE PURPOSE OF ENABLING A FUNCTIONALITY OF A USER'S DEVICE, AND THE USER'S DEVICE |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080275349A1 (en) * | 2007-05-02 | 2008-11-06 | Earlysense Ltd. | Monitoring, predicting and treating clinical episodes |
US20090005652A1 (en) * | 2007-05-07 | 2009-01-01 | Ron Kurtz | Method and system for permitting access to equipment, devices, systems, services or the like based on sleep quality analysis |
US20120179061A1 (en) * | 2009-07-16 | 2012-07-12 | Resmed Limited | Detection of sleep condition |
Family Cites Families (92)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4655213A (en) | 1983-10-06 | 1987-04-07 | New York University | Method and apparatus for the treatment of obstructive sleep apnea |
US5199424A (en) | 1987-06-26 | 1993-04-06 | Sullivan Colin E | Device for monitoring breathing during sleep and control of CPAP treatment that is patient controlled |
US5047930A (en) | 1987-06-26 | 1991-09-10 | Nicolet Instrument Corporation | Method and system for analysis of long term physiological polygraphic recordings |
US5522382A (en) | 1987-06-26 | 1996-06-04 | Rescare Limited | Device and method for treating obstructed breathing having a delay/ramp feature |
US5065756A (en) | 1987-12-22 | 1991-11-19 | New York University | Method and apparatus for the treatment of obstructive sleep apnea |
JPH0341926A (en) | 1989-07-07 | 1991-02-22 | Matsushita Electric Works Ltd | Detector for change in sleeping state and sleeping state controller |
US5239995A (en) | 1989-09-22 | 1993-08-31 | Respironics, Inc. | Sleep apnea treatment apparatus |
USRE35295E (en) | 1989-09-22 | 1996-07-16 | Respironics, Inc. | Sleep apnea treatment apparatus |
US5148802B1 (en) | 1989-09-22 | 1997-08-12 | Respironics Inc | Method and apparatus for maintaining airway patency to treat sleep apnea and other disorders |
US5632269A (en) | 1989-09-22 | 1997-05-27 | Respironics Inc. | Breathing gas delivery method and apparatus |
US5458137A (en) | 1991-06-14 | 1995-10-17 | Respironics, Inc. | Method and apparatus for controlling sleep disorder breathing |
US6629527B1 (en) | 1991-10-17 | 2003-10-07 | Respironics, Inc. | Sleep apnea treatment apparatus |
US7013892B2 (en) | 1991-11-01 | 2006-03-21 | Ric Investments, Llc | Sleep apnea treatment apparatus |
ES2146591T3 (en) | 1991-11-14 | 2000-08-16 | Univ Technologies Int | AUTOMATIC SYSTEM TO PROVIDE A CONTINUOUS POSITIVE PRESSURE IN THE NASAL RESPIRATORY TRACT. |
US5309921A (en) | 1992-02-11 | 1994-05-10 | Spectrum Medical Technologies | Apparatus and method for respiratory monitoring |
US5800360A (en) | 1992-02-11 | 1998-09-01 | Spectrum Medical Technologies, Inc. | Apparatus and method for respiratory monitoring |
US5732696A (en) | 1992-03-17 | 1998-03-31 | New York University | Polysomnograph scoring |
US5335654A (en) | 1992-05-07 | 1994-08-09 | New York University | Method and apparatus for continuous adjustment of positive airway pressure for treating obstructive sleep apnea |
US5490502A (en) | 1992-05-07 | 1996-02-13 | New York University | Method and apparatus for optimizing the continuous positive airway pressure for treating obstructive sleep apnea |
US5803066A (en) | 1992-05-07 | 1998-09-08 | New York University | Method and apparatus for optimizing the continuous positive airway pressure for treating obstructive sleep apnea |
US5593713A (en) | 1993-10-12 | 1997-01-14 | De La Luz-Martinez; Jose | Method for cooking tortillas using very low and low frequency radio waves |
EP1488743A3 (en) | 1993-11-05 | 2005-01-12 | Resmed Limited | Control of CPAP Treatment |
AUPM279393A0 (en) | 1993-12-03 | 1994-01-06 | Rescare Limited | Estimation of flow and detection of breathing in cpap treatment |
US6105575A (en) | 1994-06-03 | 2000-08-22 | Respironics, Inc. | Method and apparatus for providing positive airway pressure to a patient |
US5485851A (en) | 1994-09-21 | 1996-01-23 | Medtronic, Inc. | Method and apparatus for arousal detection |
US5551419A (en) | 1994-12-15 | 1996-09-03 | Devilbiss Health Care, Inc. | Control for CPAP apparatus |
AUPN627395A0 (en) | 1995-10-31 | 1995-11-23 | Compumedics Sleep Pty Ltd | Integrated diagnostic and therapeutic device for gas delivery to patient |
JP3737121B2 (en) | 1995-11-17 | 2006-01-18 | ニューヨーク ユニヴァーシティー | Apparatus and method for pressure and temperature waveform analysis |
US5657752A (en) | 1996-03-28 | 1997-08-19 | Airways Associates | Nasal positive airway pressure mask and method |
US6192886B1 (en) | 1996-10-17 | 2001-02-27 | Hans Rudolph, Inc. | Nasal mask |
US6070098A (en) | 1997-01-11 | 2000-05-30 | Circadian Technologies, Inc. | Method of and apparatus for evaluation and mitigation of microsleep events |
ATE383814T1 (en) | 1997-03-17 | 2008-02-15 | Vivometrics Inc | METHOD FOR RESPIRATORY WAVE FORM ANALYSIS IN RELATION TO ITS INFLUENCE ON NEUROMUSCULAR BREATHING |
US5902250A (en) | 1997-03-31 | 1999-05-11 | President And Fellows Of Harvard College | Home-based system and method for monitoring sleep state and assessing cardiorespiratory risk |
US6135106A (en) | 1997-08-22 | 2000-10-24 | Nellcor Puritan-Bennett, Inc. | CPAP pressure and flow transducer |
US6142952A (en) | 1997-10-29 | 2000-11-07 | The Board Of Regents, The University Of Texas System | Method and apparatus for detection and diagnosis of airway obstruction degree |
AUPP026997A0 (en) | 1997-11-07 | 1997-12-04 | Resmed Limited | Administration of cpap treatment pressure in presence of apnea |
US6083173A (en) | 1998-03-06 | 2000-07-04 | Research Foundation Of State University Of New York | Artificial neural network for predicting respiratory disturbances and method for developing the same |
FR2779932B1 (en) | 1998-06-18 | 2000-12-01 | Taema | DEVICE FOR DETERMINING RESPIRATORY PHASES OF THE SLEEP OF A USER |
US6171258B1 (en) | 1998-10-08 | 2001-01-09 | Sleep Solutions, Inc. | Multi-channel self-contained apparatus and method for diagnosis of sleep disorders |
US6360741B2 (en) | 1998-11-25 | 2002-03-26 | Respironics, Inc. | Pressure support system with a low leak alarm and method of using same |
AUPP783198A0 (en) | 1998-12-21 | 1999-01-21 | Resmed Limited | Determination of mask fitting pressure and correct mask fit |
US7204250B1 (en) | 1999-12-16 | 2007-04-17 | Compumedics Limited | Bio-mask |
US6477408B1 (en) | 2000-03-07 | 2002-11-05 | Northwestern University | Analysis of muscular activity in neonatal animals to screen for mutations and/or drugs that alter sleep and wake states |
US6644312B2 (en) | 2000-03-07 | 2003-11-11 | Resmed Limited | Determining suitable ventilator settings for patients with alveolar hypoventilation during sleep |
US6371112B1 (en) | 2000-05-22 | 2002-04-16 | Noam Bibi | Device, system and method for preventing collapse of the upper airway |
ES2250263T3 (en) | 2000-06-14 | 2006-04-16 | FISHER & PAYKEL HEALTHCARE LIMITED | NASAL MASK. |
US6605038B1 (en) | 2000-06-16 | 2003-08-12 | Bodymedia, Inc. | System for monitoring health, wellness and fitness |
DE10031079A1 (en) | 2000-06-30 | 2002-02-07 | Map Gmbh | Measuring patient breathing and state, correlates present respiration signals with prior reference measurements, to adjust CPAP therapy pressure accordingly |
US6349724B1 (en) | 2000-07-05 | 2002-02-26 | Compumedics Sleep Pty. Ltd. | Dual-pressure blower for positive air pressure device |
US6856829B2 (en) | 2000-09-07 | 2005-02-15 | Denso Corporation | Method for detecting physiological condition of sleeping patient based on analysis of pulse waves |
SE0003531D0 (en) | 2000-10-02 | 2000-10-02 | Breas Medical Ab | Auto CPAP |
US6581595B1 (en) | 2000-11-14 | 2003-06-24 | Sensormedics Corporation | Positive airway pressure device with indirect calorimetry system |
CA2430775A1 (en) | 2000-12-11 | 2002-06-20 | Resmed Ltd. | Method and apparatus for stroke patient treatment |
US6641542B2 (en) | 2001-04-30 | 2003-11-04 | Medtronic, Inc. | Method and apparatus to detect and treat sleep respiratory events |
AU2002308423B2 (en) | 2001-05-23 | 2007-11-01 | Resmed Limited | Ventilator patient synchronization |
US6805668B1 (en) | 2001-06-26 | 2004-10-19 | Cadwell Industries, Inc. | System and method for processing patient polysomnograph data utilizing multiple neural network processing |
EP1418968A4 (en) | 2001-07-19 | 2009-11-18 | Resmed Ltd | Pressure support ventilation of patients |
FR2833496B1 (en) | 2001-12-14 | 2004-02-13 | Ela Medical Sa | ACTIVE MEDICAL DEVICE COMPRISING IMPROVED MEANS OF DIAGNOSING SLEEP APNEA SYNDROME |
IL147502A0 (en) | 2002-01-07 | 2002-08-14 | Widemed Ltd | Self-adaptive system, for the analysis of biomedical signals of a patient |
DE10226783A1 (en) | 2002-06-15 | 2004-01-08 | Gottlieb Weinmann Geräte für Medizin und Arbeitsschutz GmbH & Co. | Process for controlling a ventilator and device for ventilation with control |
KR20050072435A (en) | 2002-10-09 | 2005-07-11 | 컴퓨메딕스 리미티드 | Method and apparatus for maintaining and monitoring sleep quality during therapeutic treatments |
US7189204B2 (en) | 2002-12-04 | 2007-03-13 | Cardiac Pacemakers, Inc. | Sleep detection using an adjustable threshold |
US7207947B2 (en) | 2003-01-10 | 2007-04-24 | Pacesetter, Inc. | System and method for detecting circadian states using an implantable medical device |
US20060249149A1 (en) | 2003-03-17 | 2006-11-09 | Meier Joerg | Method and arrangement for the tiration of physiological measuring signals in conjunction with the observation of a patient in terms of sleep-related respiratory problems |
US6988994B2 (en) | 2003-08-14 | 2006-01-24 | New York University | Positive airway pressure system and method for treatment of sleeping disorder in patient |
US7896812B2 (en) | 2003-08-14 | 2011-03-01 | New York University | System and method for diagnosis and treatment of a breathing pattern of a patient |
US7887493B2 (en) | 2003-09-18 | 2011-02-15 | Cardiac Pacemakers, Inc. | Implantable device employing movement sensing for detecting sleep-related disorders |
US7572225B2 (en) | 2003-09-18 | 2009-08-11 | Cardiac Pacemakers, Inc. | Sleep logbook |
EP1670547B1 (en) | 2003-08-18 | 2008-11-12 | Cardiac Pacemakers, Inc. | Patient monitoring system |
US8002553B2 (en) | 2003-08-18 | 2011-08-23 | Cardiac Pacemakers, Inc. | Sleep quality data collection and evaluation |
US8192376B2 (en) | 2003-08-18 | 2012-06-05 | Cardiac Pacemakers, Inc. | Sleep state classification |
US7662101B2 (en) | 2003-09-18 | 2010-02-16 | Cardiac Pacemakers, Inc. | Therapy control based on cardiopulmonary status |
US7469697B2 (en) | 2003-09-18 | 2008-12-30 | Cardiac Pacemakers, Inc. | Feedback system and method for sleep disordered breathing therapy |
JP3733133B2 (en) | 2003-10-14 | 2006-01-11 | 三洋電機株式会社 | Sleep state estimation device |
EP1680166B1 (en) | 2003-10-17 | 2019-09-18 | ResMed Pty Ltd | Apparatus for heart failure treatment |
US7041049B1 (en) | 2003-11-21 | 2006-05-09 | First Principles, Inc. | Sleep guidance system and related methods |
JP4753881B2 (en) | 2004-01-16 | 2011-08-24 | コンプメディクス リミテッド | Apparatus and signal processing method for monitoring, detecting and classifying sleep disordered breathing derived from electrocardiogram |
US20050209512A1 (en) | 2004-03-16 | 2005-09-22 | Heruth Kenneth T | Detecting sleep |
US7717848B2 (en) | 2004-03-16 | 2010-05-18 | Medtronic, Inc. | Collecting sleep quality information via a medical device |
US7276031B2 (en) | 2004-05-12 | 2007-10-02 | New York University | System and method for classifying patient's breathing using artificial neural network |
JP3976752B2 (en) | 2004-07-07 | 2007-09-19 | 三洋電機株式会社 | Sleep state estimation apparatus and program |
EP1893264B1 (en) | 2005-06-14 | 2016-02-24 | ResMed Limited | Acclimatization therapy for first time cpap and niv users |
US8083682B2 (en) | 2006-07-19 | 2011-12-27 | Cardiac Pacemakers, Inc. | Sleep state detection |
WO2010021556A1 (en) * | 2008-08-19 | 2010-02-25 | Fisher & Paykel Healthcare Limited | Breathing transition detection |
CA2819647C (en) | 2010-12-03 | 2019-11-26 | Fisher & Paykel Healthcare Limited | System, apparatus and methods for supplying gases |
WO2014007659A1 (en) | 2012-02-15 | 2014-01-09 | Fisher & Paykel Healthcare Limited | System, apparatus and methods for supplying gases |
EP2818111B1 (en) * | 2012-02-24 | 2020-02-05 | Toyota Jidosha Kabushiki Kaisha | Sleep quality estimation device and sleep quality estimation method. |
AU2013318046B2 (en) | 2012-09-19 | 2016-07-21 | Resmed Sensor Technologies Limited | System and method for determining sleep stage |
SG11201602038PA (en) | 2013-09-17 | 2016-04-28 | Fisher & Paykel Healthcare Ltd | Valve with internal member |
WO2015052681A1 (en) | 2013-10-11 | 2015-04-16 | Fisher & Paykel Healthcare Limited | Hme and compact breathing apparatus |
US9655559B2 (en) * | 2014-01-03 | 2017-05-23 | Vital Connect, Inc. | Automated sleep staging using wearable sensors |
WO2018009078A1 (en) | 2016-07-06 | 2018-01-11 | Fisher & Paykel Healthcare Limited | A bilevel respiratory therapy system, controller and method |
-
2016
- 2016-12-08 US US15/373,344 patent/US11298074B2/en active Active
-
2022
- 2022-03-02 US US17/653,255 patent/US20220313155A1/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080275349A1 (en) * | 2007-05-02 | 2008-11-06 | Earlysense Ltd. | Monitoring, predicting and treating clinical episodes |
US20090005652A1 (en) * | 2007-05-07 | 2009-01-01 | Ron Kurtz | Method and system for permitting access to equipment, devices, systems, services or the like based on sleep quality analysis |
US20120179061A1 (en) * | 2009-07-16 | 2012-07-12 | Resmed Limited | Detection of sleep condition |
Also Published As
Publication number | Publication date |
---|---|
US20170196500A1 (en) | 2017-07-13 |
US11298074B2 (en) | 2022-04-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20220313155A1 (en) | Flow-based sleep stage determination | |
US11040159B2 (en) | Distinguishing between central and obstructive sleep apnea | |
US11751817B2 (en) | Breathing disorder identification, characterization and diagnosis methods, devices and systems | |
US20160045161A1 (en) | Mask and method for breathing disorder identification, characterization and/or diagnosis | |
US20190015614A1 (en) | Systems, devices, and methodologies to provide protective and personalized ventilation | |
US11612708B2 (en) | Method and apparatus for determining and/or predicting sleep and respiratory behaviours for management of airway pressure | |
KR20030066632A (en) | Method and apparatus for cpap using a neural network | |
CN103221086B (en) | Fat hypoventilation syndrome disposal system | |
JP2008501425A (en) | Automatic titration system and method for continuous positive pressure breathing | |
WO2018120643A1 (en) | Physiological monitoring result feedback method and device | |
US11676713B2 (en) | Data processing system for classifying keyed data representing inhaler device operation | |
Ozdemir et al. | A time-series approach to predict obstructive sleep apnea (OSA) Episodes | |
JP6557314B2 (en) | Lung health monitoring system | |
US20160287822A1 (en) | Systems and Methods for Detection of Respiratory Events Using Pattern Recognition | |
Eedara et al. | An algorithm for automatic respiratory state classifications using tracheal sound analysis | |
US12016672B2 (en) | Method and system for indicating obstructive sleep apnea during wakefulness | |
Castro et al. | Real-time identification of respiratory movements through a microphone | |
US20210282736A1 (en) | Respiration rate detection metholody for nebulizers | |
WO2017141057A1 (en) | Communication apparatus, method and computer program | |
Unger | Analysis of Respiratory Time Series Data for Breathing Discomfort Detection Prior to Sleep Onset During APAP Therapy | |
Premasiri et al. | A multi-sensor data fusion approach for sleep apnea monitoring using neural networks | |
Chiu et al. | Two-Stage Segment-Based Acoustic Approach for Fast Glottis Identification During Endotracheal Intubation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STPP | Information on status: patent application and granting procedure in general |
Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: NON FINAL ACTION MAILED |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER |