GB2600209A

GB2600209A - System and method for improving sleep

Info

Publication number: GB2600209A
Application number: GB2109393.5A
Authority: GB
Inventors: Robyn Suer Francesca
Original assignee: Individual
Current assignee: Individual
Priority date: 2020-06-29
Filing date: 2021-06-29
Publication date: 2022-04-27
Also published as: GB202009898D0; GB202109393D0

Abstract

A method and system for improving the quality of sleep; the method comprises monitoring the actual sleep of a sleeping person; providing a model of ideal sleep for the person; monitoring at least one parameter of a sleep environment; comparing the actual sleep with the model of ideal sleep; identifying a difference between the actual sleep and the ideal sleep; and based n the difference, determine whether to modify a parameter of the sleep environment. Additionally disclosed is a system for providing a model of ideal sleep of a person over a plurality of instances to gather data; analysis of the gathered data to extract features therefrom; classifying sleep stages and or transitions from the sleep data; and generating a model of ideal sleep for the person.

Description

SYSTEM AND METHOD FOR IMPROVING SLEEP

Field

This invention relates to improving sleep, and in particular relates to detecting microclimate induced abnormal sleep behaviour, and providing a system to mitigate such behaviour, for example using smart home technology actuation.

Background

With only one third of adults across developed nations obtaining the recommended eight hours of sleep, insufficient sleep quality is a problem that has adverse consequences in a huge variety of aspects, including health, economic and social issues. For example, lack of sleep can lead to increased likelihood of disease and mental health illnesses, lower productivity levels estimated to cost the UK around £40 billion annually, and emotional instability, dishonesty and aggressive behaviour, all fostering unfavourable social interactions.

Sleep may be influenced by fixed parameters such as gender, age and genetics, by daily contextual factors such as exercise level, sleep routine and alcohol intake that may require behaviour change to alter, and by the bedroom microclimate including but not limited to light, temperature, humidity and sound level.

Given the rising trend in automated smart home technologies, there is an opportunity for night-time microclimate actuation which might improve sleep.

Summary of Invention

The invention aims to build a system to identify microclimate induced abnormal or undesirable sleep behaviour in real time based on physiological, contextual and local microclimate data which may be monitored by existing consumer smart home and sleep monitoring technologies and/or wearables. It then aims to derive the cause of the abnormal or undesirable events so that the microclimate may be adapted in real time, potentially to prevent the abnormal event from occurring, for example relevant smart home technologies may be integrated into the system.

Thus according to the present invention, there is provided a method of improving sleep comprising: monitoring the actual sleep of a sleeping person; providing a model of ideal sleep for the person; monitoring at least one parameter of a sleep environment; comparing the actual sleep with the model of ideal sleep; identifying a difference between the actual sleep and the ideal sleep.

Preferably, the method includes, based on the identified difference, modifying the or each parameter of the sleep environment. Thus it may be possible to encourage good or ideal sleep. For example, if it is determined that a sleep transition is occurring which is undesirable, such as the person waking up after an insufficient time period spent asleep or spent in a particular sleep stage, the method may modify the sleep environment in order to prevent such a transition.

Preferably, the method includes, identifying a parameter of the sleep environment which has changed or which is outside a predetermined range of values and which corresponds to or occurs at the same time as the difference between the actual and ideal sleep, and modifying the identified parameter. Thus the method can identify and correct a cause of an undesirable change in sleep.

In one example, the model may include the time period spent in each of a plurality of sleep stages, and/or the time stamp of one or more transitions between stages, and/or an ideal or preferred ratio of various sleep stages during one period of sleep, such as one night. Thus preferably the method includes updating the model of ideal sleep dependent upon the monitored actual sleep. Thus for example the time periods, time stamps and/or the ratio of different sleep stages in the model may be modified depending upon the time periods, time stamps and/or ratio of sleep stages already monitored during a sleep period. The model may also include the total time spent asleep.

Some variance of sleep from the model may not be undesirable for ideal or good sleep. Therefore preferably, the method includes determining whether the identified difference is an undesirable difference, and modifying the or each parameter only if the difference is undesirable. For example, if a sleep stage or transition occurs within a predetermined range of variance from the model, then the difference is not determined to be undesirable, and the or each parameter may not be modified.

Preferably, monitoring the actual sleep comprises identifying or predicting sleep transitions and/or sleep stages, such as transitions between REM and NREM stages, and between NREM stages, and in and out of sleep.

Preferably, providing a model of ideal sleep for the person includes sleep transitions and/or sleep stages, such as transitions between REM and NREM stages, and between NREM stages, and in or out of sleep.

Conveniently, monitoring the actual sleep comprises identifying the time spent asleep, and/or the time spent in each sleep stage, and/or the time stamp of the beginning or end of each sleep stage.

Preferably, monitoring the at least one parameter includes monitoring at least one of the temperature, humidity level, ambient noise level, light intensity, light wavelength, and/or light quality.

Preferably, when there is more than one parameter of the sleep environment being monitored, the method includes selecting which parameter or parameters to modify, and preferably, modifying the or each parameter of the sleep environment which is determined as having an effect on the actual sleep of the sleeping person, and/or which is determined to be causing an abnormal or undesirable difference.

Preferably, modifying the or each parameter of the sleep environment is effected by one or more smart home apparatuses.

Preferably, the parameter of the sleep environment is selected by comparing the value of the or each parameter with an ideal or predetermined value or range of values, and determining that the parameter is different from the ideal value or range of values; or determining that the parameter has changed by more than a predetermined amount.

Preferably, the ideal value or range of values may differ based upon the transition between sleep stages and/or the sleep stage in which the sleeping person is.

Conveniently, if the actual sleep data indicates that a sleep stage transition is occurring which is does not correlate with the model, and/or which is determined (from the model) to be undesirable, the or each parameter of the sleep environment is adjusted with the aim to prevent the incorrect and/or undesirable sleep stage transition.

Preferably, monitoring the actual sleep may comprise monitoring a physical or biometric parameter of a sleeping person, such as heart rate, respiratory rate, movement, core body temperature, or variability of fluctuations of the physical parameters, such as heart rate variability, HRV, or respiratory rate variability, RRV.

Monitoring the actual sleep may be done with or without contact with the person. Conveniently, the heart rate, respiratory rate, heart rate variability, HRV, movement and/or respiratory rate variability, RRV, may be detected using an under-mattress device.

Alternatively, a wearable device may be used.

The sleep stage or sleep transition may be derived from the monitoring of such physical or biometric features.

Advantageously, the heart rate and/or heart rate variability, HRV, and other biometric features may be derived from a ballistocardiograph, BCG, signal. Heart rate or other biometric features may be measured in various other ways.

Preferably, providing the model of ideal sleep may comprise the steps of: monitoring the actual sleep of a sleeping person over a plurality of instances of sleep to gather sleep data; analysing the sleep data to extract features therefrom; classifying sleep stages and sleep stage transitions from the features from the sleep data; 30 and generating a model of ideal sleep for the person.

Conveniently, the sleep data may be compared against a reference data set to aid in classification of sleep stages.

Preferably, the model of ideal sleep may be adjusted based upon parameters which relate to the person's attributes. For example, this may include the person's age or weight. Advantageously, the model may be modified depending upon variables relating to the experience of the person during a preceding time period such as a day, which may include level and/or type of activity, consumption of food or drink, in particular alcohol and/or caffeine, and the time at which sleep began and/or finished.

For example, different models may be provided depending upon an experience or category of experience, or a person's attributes, having a particular value or falling into a particular range.

Conveniently, the variables relating to the experiences or attributes may be collected by one or more devices such as wearable devices.

Preferably, the model includes identification of the person's transitions between sleep stages.

Conveniently, the person's transitions between sleep stages are used to generate the model of ideal sleep.

Alternatively, the model of ideal sleep may be selected from a database of reference sleep data, for example reference sleep compositions for a period of sleep such as a night, which data may be selected dependent upon a parameter such as a fixed parameter or attribute relating to the person, such as age or gender.

The model may then be modified, or a different model may be selected, before each period of sleep in dependence upon variable data such as gathered contextual data relating to the person's experience during the time period preceding the period of sleep, The model may then be adjusted in real time during the period of sleep using actual data from the sleeping person, for example by the method described above.

Sleep compositions may comprise time stamps for each sleep stage transition.

According to another aspect of the invention, there is provided a method of providing a model of an ideal sleep pattern for a person, comprising: monitoring the actual sleep of a sleeping person over a plurality of instances of sleep to gather sleep data; analysing the sleep data to extract features therefrom; classifying sleep stages and sleep stage transitions from the features from the sleep data; and generating a model of ideal sleep for the person.

The model may be adjusted by repeating the steps of gathering and analysing the sleep data.

Brief Description of Figures

Embodiments will now be described, by way of example only, and with reference to the accompanying drawings, in which: Figure 1 illustrates an example of sleep stages, as known in the art; Figure 2 illustrates a schematic illustration of the system; Figure 3 illustrates an example of a sleep monitoring device; Figure 4 illustrates an example of a ballistocardiograph, 'BCG', signal; Figure 5 illustrates an example of sleep stages, and includes labelled transitions between sleep stages; Figure 6 illustrates an example of feature selection; and Figure 7 shows an illustration of system to improve sleep quality via optimisation of the local microclimate.

Specific Description

A network of monitoring devices can be provided that track sleep quality, the bedroom microclimate and daily contextual factors known to influence sleep. By integrating two time series predictive models constructed and evaluated using methods of signal processing and machine learning, the system identifies microclimate induced abnormal sleep behaviour so that the bedroom microclimate can be optimised in real time to improve sleep quality.

Sleep quality is reflected in sleep composition, formed by several cycles that comprise distinct stages. During NREM (non-rapid eye movement) stages, heart rate, respiratory rate and brainwave activity all slow down, whereas during REM (rapid eye movement) where dreaming typically occurs, these biometrics all increase. Therefore, by analysing biometric features it is possible to classify sleep stage and assess sleep quality. This can be done in sleep science and some consumer sleep assessment technologies. Figure 1 shows an example of sleep stages.

In order to carry out the method, two predictive models are built, one to classify current sleep (Model 1) and another to predict sleep in an ideal microclimate in a specific contextual circumstance (Model 2). Comparisons between the two models can be used to identify microclimate induced abnormal sleep behaviour. By cross-referencing the timestamp of abnormal behaviour with microclimate parameters, the cause can be identified and then rectified with microclimate actuation devices. Additionally, to prevent abnormal sleep behaviour ahead of it occurring, sleep stage transition classification can be used in additional to sleep stage classification. Figure 2 shows an example of the system.

In order to monitor sleep, a monitoring device such as the Emfit QS+Active under-mattress device can be used. This detects a 'ballistocardiograph' (or 'BCG') signal, a measurement of force. To monitor the bedroom microclimate, various microclimate sensors can be used, for example connected to a processor such as a Raspberry Pi.

Additionally, a diary can be kept to record relevant daily contextual data. Figure 3 shows an example of a sleep monitoring device.

From the BCG signal body movement and heart/respiratory pulsations can be derived.

Heart rate, heart rate variability, activity and activity level variability can all be calculated by extracting relevant peaks & waveforms from the signal. Figure 4 shows an example of a BCG signal, with an example section extracted.

Feature selection can be used to identify features relevant to the labelled data. Data was labelled with four sleep stages and the twelve possible transition variations with transitions labelled as the 30 second period preceding a change of sleep stage. Figure 5 shows an example of sleep stages, and includes labelled transitions.

Feature selection can be conducted with biometric features extracted from the BCG signal for Model l's input and sleep composition features extracted from Model l's output for Model 2's input. Figure 6 shows an example of feature selection.

Recurrent Neural Networks can be used, a class of Neural Networks that recognise patterns in sequential data. The models may be run with a plurality of nights of data, for example with 20 nights or more of data, split into a training, validation and test set. With 20 nights of data, models 1 & 2 were found to achieve accuracies of 49.3% & 49.4% on the test sets respectively. It's important to note therefore that to build accurate models much more data is necessary, especially for classifying transitions that form such a small portion of the data.

Using insights from this process, a system was proposed to improve sleep quality via optimisation of the local microclimate. An example of such a system is shown in Figure 7.

The bedroom microclimate is monitored by various smart home devices. Sleep quality is monitored by an under-mattress BCG sensor from which biometric features are extracted, used as input for Model 1 for real time sleep stage/transition classification.

Daily contextual factors are monitored by calendar data, wearables etc. The system builds a selection of predictive models from nights categorised by contextual data. The contextual data from this day is used to select the model suitable for this night: Model 2. This model bases sleep stage/transition classification on sleep composition in an ideal microclimate given a particular contextual circumstance. It takes sleep composition feature input from Model l's output.

If a transition is predicted, predictions are compared and if discrepancies are above a set threshold, this indicates a microclimate induced abnormal transition and the microclimate is checked for fluctuations that indicate the cause. Once communicated to relevant smart home devices, the abnormal transition can be stopped from proceeding via corrective actuation.

Any method features may be applied to or provided as features of the system, and vice versa.

While the invention has been illustrated and described in detail in the drawings and preceding description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments.

Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. Each feature of the disclosed embodiments may be replaced by alternative features serving the same, equivalent or similar purpose, unless stated otherwise. Therefore, unless stated otherwise, each feature disclosed is one example of a generic series of equivalent or similar features.

In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. Any reference signs in the claims should not be construed as limiting the scope.

Claims

Claims 1. A method of improving sleep including: monitoring the actual sleep of a sleeping person; providing a model of ideal sleep for the person; monitoring at least one parameter of a sleep environment; comparing the actual sleep with the model of ideal sleep; identifying a difference between the actual sleep and the ideal sleep; and based on the identified difference, determining whether to modify at least one parameter of the sleep environment, and if so, modifying the or each parameter.
2. The method of claim 1, wherein determining whether to modify at least one parameter of the sleep environment includes determining whether the identified difference is undesirable, and/or determining whether a parameter has changed more than a predetermined amount or has a value which lies outside of a predetermined range.
3. The method of claim 1 or 2, including updating the model of ideal sleep dependent upon the monitored actual sleep.
4. The method of claim 1, 2 or 3,wherein monitoring the actual sleep includes identifying one or more sleep transitions and/or sleep stages, such as REM and NREM stages, and transitions between REM and NREM stages, between different NREM stages, and in and out of sleep.
5. The method of any previous claim, wherein providing a model of ideal sleep for the person includes one or more sleep transitions and/or sleep stages, such as REM and NREM stages, and transitions between REM and NREM stages, between NREM stages, and in and out of sleep.
6. The method of any previous claim, wherein monitoring the actual sleep includes identifying the time or time period asleep, and/or the time or time period in each sleep stage.
7. The method of any previous claim, wherein monitoring the at least one parameter includes monitoring the temperature, humidity level, ambient noise level, light intensity, light wavelength, and/or light quality.
8. The method of any previous claim, wherein when there is more than one parameter of the sleep environment being monitored, the method includes selecting which parameter to modify.
9. The method of any previous claim, further including the step of selecting one or more parameters of the sleep environment which is or are determined as having an effect on the actual sleep of the sleeping person, and modifying the or each selected parameter.
10. The method of any previous claim, wherein modifying the or each parameter of the sleep environment is effected by one or more smart home apparatuses.
11. The method of any previous claim, wherein the parameter of the sleep environment is selected by comparing the value of the or each parameter with an ideal or predetermined value or range of values, and determining that the parameter is different from the ideal or predetermined value or range of values; or determining that the parameter has changed by more than a predetermined amount.
12. The method of claim 11, wherein the ideal or predetermined value or range of values of the one or more parameters of the sleep environment may differ based upon the transition between sleep stages and/or the sleep stage in which the sleeping person is.
13. The method of any previous claim, wherein if the actual sleep data indicates that a sleep stage transition is occurring which does not correlate with the model and/or is determined to be undesirable, the or each parameter of the sleep environment is adjusted.
14. The method of any previous claim, wherein monitoring the actual sleep may include monitoring a physical parameter of a sleeping person, such as heart rate, respiratory rate, movement, core body temperature, or variability of fluctuations of the physical parameters, such as heart rate variability, HRV, or respiratory rate variability, RRV.
15. The method of claim 14, wherein biometric features such as the heart rate, respiratory rate, heart rate variability, HRV, movement and/or respiratory rate variability, RRV, may be detected using an under-mattress device and/or a wearable device.
16. The method of claim 14 or 15, wherein the heart rate and/or heart rate variability, HRV, are derived from a ballistocardiograph, BCG, signal.
17. The method of any previous claim, wherein providing the model of ideal sleep includes the steps of: monitoring the actual sleep of a sleeping person over a plurality of instances of sleep to gather sleep data; analysing the sleep data to extract features therefrom; classifying sleep stages and/or sleep stage transitions from the features from the sleep data; and generating a model of ideal sleep for the person.
18. The method of claim 17, wherein the sleep data is compared against a reference data set to aid in classification of sleep stages and/or transitions.
19. The method of claims 17 or 18, wherein the model of ideal sleep is adjusted based upon parameters which relate to the person's attributes, which may include age or weight, and/or wherein the model is modified depending upon variables relating to the experience of the person during a preceding time period or time between sleep periods such as a day, which may include level and/or type of activity, consumption of food or drink, in particular alcohol and/or caffeine, and the time at which sleep began and/or finished.
20. The method of claim 19, wherein the variables relating to the experiences or attributes are collected by one or more devices such as wearable devices.
21. A method for providing a model of an ideal sleep pattern for a person, including: monitoring the actual sleep of a sleeping person over a plurality of instances of sleep to gather sleep data; analysing the sleep data to extract features therefrom; classifying sleep stages and/or sleep stage transitions from the features from the sleep data; and generating a model of ideal sleep for the person, and optionally repeating the steps of monitoring, analysing and classifying the sleep, and modifying the model accordingly.
22. A system configured to carry out the method of any preceding claim.
23. A system for improving sleep including: a sleep analysis arrangement configured to monitor the actual sleep of a sleeping person; a modelling arrangement configured to provide a model of ideal sleep for the person; an environmental monitoring arrangement configured to monitor at least one parameter of a sleep environment; a comparator configured to compare the actual sleep with the model of ideal sleep; a determination arrangement configured to identify a difference between the actual sleep and the ideal sleep; and an environmental modification arrangement which is configured to modify the or each parameter of the sleep environment, based on the identified difference.
24. A system for providing a model of an ideal sleep pattern for a person, including: a sleep analysis arrangement configured to monitor the actual sleep of a sleeping person over a plurality of instances of sleep to gather sleep data; an analysis arrangement configured to analyse the sleep data to extract features 30 therefrom; a classification arrangement configured to classify sleep stages and/or sleep stage transitions from the features from the sleep data; and a generating arrangement configured to generate a model of ideal sleep for the person.
25. A non-transitory computer readable medium having stored thereon software instructions that, when executed by a processor, cause the processor to carry out the steps of any one of method claims 1 to 22.