An improved method, system, and device for delivering stimulation during sleep

Abstract

Described herein is a method for providing personalised stimulation to a user during a sleep period comprising the steps of: receiving one or more signals indicative of the user's sleep activity during a sleep period, extracting one or more features of the one or more signals, determining a stimulation response including determining whether to vary one or more stimulation response parameters based on the extracted one or more features of the one or more signals, and delivering a stimulation response to the user based on the determined stimulation response including varying the stimulation response by varying one or more stimulation response parameters.

Description

AN IMPROVED METHOD, SYSTEM, AND DEVICE FOR DELIVERING STIMULATION DURING SLEEPTECHNICAL FIELD

[0001] The embodiments described herein broadly relate to a method, system, and device for delivering stimulation to a user during sleep.BACKGROUND

[0002] Existing techniques in the domain of phase-targeted auditory stimulation during user sleep typically rely on static timing or volume control using a threshold measure for volume adjustment. Existing techniques currently target a fixed point in the slow-wave oscillation for auditory stimulation. For example, prior systems have used phase-locked loops (PLLs) or similar estimators to target the up-phase of slow-wave sleep, usually at a fixed delay (e.g., 20-30 degrees from the peak) and with a fixed stimulus duration (typically 50 ms), usually with pink-noise or similar broad- spectrum sound.

[0003] Further, existing headbands that provide stimulation during user sleep typically provide the electronics on the front of the headband device such that they rest on the user’s forehead or top of head. The obvious approach is to place the electronics on the front as that is the location of electrodes, speaker, and heart rate sensor, and is the obvious place to put a bulky piece as it will interfere with sleep less in most positions.SUMMARY

[0004] It is desirable to provide a method and device for delivering sleep stimulation. Additionally or alternatively, it is desirable to provide the industry with a useful choice.

[0005] In one aspect, there is provided a method for providing personalised stimulation to a user during a sleep period comprising the steps of: receiving one or more signals indicative of the user’s sleep activity during a sleep period, extracting one or more features of the one or more signals, determining a stimulation response including determining whether to vary one or more stimulation response parameters based on the extracted one or more features ofthe one or more signals, and delivering a stimulation response to the user based on the determined stimulation response including varying the stimulation response by varying one or more stimulation response parameters.

[0006] In some embodiments, the delivered stimulation response improves the restorative function of the sleep period, optionally during deep sleep.

[0007] In some embodiments, the method further comprises the step of sensing the user’s response to the delivered stimulation.

[0008] In some embodiments, the step of receiving one or more signals indicative of the user’s sleep activity is based on the user’s sensed response to the delivered stimulation.

[0009] In some embodiments, the method further includes the step of evaluating the user’s sensed response to the delivered stimulation.

[0010] In some embodiments, the step of evaluating the user’s sensed response to the delivered stimulation includes assigning a reward score for each delivered stimulation response.

[0011] In some embodiments, the method further comprises the step of updating a model based on the user’s evaluated response to the delivered stimulation.

[0012] In some embodiments, the step of determining a stimulation response is further based on one or more outputs of the updated model.

[0013] In some embodiments, the method further comprises the step of filtering the one or more signals for timing the stimulation response delivery to the user.

[0014] In some embodiments, the one or more stimulation response parameters includes one or more of:• volume• sound frequency• timing• time duration

[0015] In some embodiments, the one or more stimulation parameters are or comprises volume and sound frequency.

[0016] In some embodiments, the one or more stimulation parameters are or comprises timing relative to a phase angle of a slow-wave, volume, and sound frequency.

[0017] In some embodiments, the one or more signals includes one or more EEG signals of the user.

[0018] In some embodiments, the one or more extracted features of the one or more EEG signals includes one or more of:• delta power• slow-wave counts• spectral activity in alpha, theta, and / or sigma bands• alpha intrusion• peak-to-peak amplitude• changes in spectral power over time• wave slope• inter-peak duration.

[0019] In some embodiments, the one or more signals includes one or more signals indicative of the user’s motion.

[0020] In some embodiments, the one or more extracted features of the one or more signals indicative of the user’s motion includes one or more of:• accelerometer-based movement detection• posture shift• one or more stillness patterns

[0021] In some embodiments, the one or more signals includes one or more signals indicative of the user’s heart activity.

[0022] In some embodiments, the one or more extracted features of the one or more signals indicative of the user’s heart activity includes one or more of:• heart rate• heart rate variability• respiration rate• oxygen saturation

[0023] In some embodiments, the one or more signals includes one or more signals indicative of the user’s one or more respiratory events.

[0024] In some embodiments, the step of sensing the user’s response to the delivered stimulation includes sensing for increased electrical activity in the user’s brain.

[0025] In some embodiments, the step of sensing the user’s response to the delivered stimulation includes sensing for increased slow-wave activity in the user’s brain prompted by the delivered stimulation.

[0026] In another aspect, there is provided a device for providing personalised stimulation to a user during a sleep period comprising one or more controllers configured to: receive one or more signals indicative of the user’s sleep activity during a sleep period, extract one or more features of the one or more signals, determine a stimulation response including determining whether to vary one or more stimulation response parameters based on the extracted one or more features of the one or more signals, and deliver a stimulation response to the user based on the determined stimulation response including varying the stimulation response by varying one or more stimulation response parameters.

[0027] In some embodiments, the device is user wearable.

[0028] In some embodiments, the device is a headband.

[0029] In some embodiments, the delivered stimulation response improves the restorative function of the sleep period, optionally during deep sleep.

[0030] In some embodiments, the one or more controllers are further configured to sense the user’s response to the delivered stimulation.

[0031] In some embodiments, receiving one or more signals indicative of the user’s sleep activity is based on the user’s sensed response to the delivered stimulation.

[0032] In some embodiments, the one or more controllers are further configured to evaluate the user’s sensed response to the delivered stimulation.

[0033] In some embodiments, evaluating the user’s sensed response to the delivered stimulation includes assigning a reward score for each delivered stimulation response.

[0034] In some embodiments, the one or more controllers are further configured to update a model based on the user’s evaluated response to the delivered stimulation.

[0035] In some embodiments, determining a stimulation response is further based on one or more outputs of the updated model.

[0036] In some embodiments, the device further comprises a Kalman filter configured to filter the one or more signals for timing the stimulation response delivery to the user.

[0037] In some embodiments, the one or more stimulation response parameters includes one or more of:• volume• sound frequency• timing• time duration

[0038] In some embodiments, the one or more stimulation parameters are or comprises volume and sound frequency.

[0039] In some embodiments, the one or more stimulation parameters are or comprises timing relative to a phase angle of a slow-wave, volume, and sound frequency.

[0040] In some embodiments, the one or more signals includes one or more EEG signals of the user.

[0041] In some embodiments, the one or more extracted features of the one or more EEG signals includes one or more of:• delta power• slow-wave counts• spectral activity in alpha, theta, and / or sigma bands• alpha intrusion• peak-to-peak amplitude• changes in spectral power over time• wave slope• inter-peak duration.

[0042] In some embodiments, the one or more signals includes one or more signals indicative of the user’s motion.

[0043] In some embodiments, the one or more extracted features of the one or more signals indicative of the user’s motion includes one or more of:• accelerometer-based movement detection• posture shift• one or more stillness patterns

[0044] In some embodiments, the one or more signals includes one or more signals indicative of the user’s heart activity.

[0045] In some embodiments, the one or more extracted features of the one or more signals indicative of the user’s heart activity includes one or more of:• heart rate• heart rate variability• respiration rate• oxygen saturation

[0046] In some embodiments, the one or more signals includes one or more signals indicative of the user’s one or more respiratory events.

[0047] In some embodiments, sensing the user’s response to the delivered stimulation includes monitoring for increased electrical activity in the user’s brain.

[0048] In some embodiments, sensing the user’s response to the delivered stimulation includes monitoring for increased slow-wave activity in the user’s brain prompted by the delivered stimulation.

[0049] In another aspect, there is provided a headband device comprising: a first interface portion to interface with a rear side of a user’s head, a first flexible arm physically coupled to the first interface portion, a second interface portion to interface with a front side of the user’s head, the second interface portion located on the first flexible arm, and wherein the flexible arm includes a flexible printed circuit or part thereof to electrically connect the first interface portion with the second interface portion.

[0050] In some embodiments, the first interface portion is larger than the second interface portion.

[0051] In some embodiments, the first interface portion comprises a battery.

[0052] In some embodiments, the first and second interface portions are flexible.

[0053] In some embodiments, the first interface portion comprises one or more printed circuit boards.

[0054] In some embodiments, each printed circuit board is contained within a respective housing.

[0055] In some embodiments, the first interface portion comprises one or more physical control housings.

[0056] In some embodiments, gaps are provided between the housings to provide flex points.

[0057] In some embodiments, the device further comprises a second flexible arm.

[0058] In some embodiments, the devices further comprises one or more flexible springs that connects between the first interface portion and a respective flexible arm.

[0059] In some embodiments, the device further comprises a removable and / or washable cover.BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Fig. 1 shows a general method embodiment.

[0061] Fig. 2 shows a general system embodiment.

[0062] Fig. 3 shows a first exemplary embodiment.

[0063] Fig. 4 shows an ERP (Event-Related Potential) response graph.

[0064] Fig. 5A shows an ERP (Event-Related Potential) response graph.

[0065] Fig. 5B shows a polar graph.

[0066] Figs. 6A-C each show an ERP (Event-Related Potential) response graph.

[0067] Fig. 7 shows a chart that displays the sound frequency to power component.

[0068] Figs. 8A-N show views of a headband device embodiment.

[0069] Figs. 9A-B show views of the printed circuit boards of the first interface portion.

[0070] Figs. 10A-B show views of the second interface portion.

[0071] Figs. 11 A-B show view of the flexible circuit board to electrically connect the first interface portion to the second interface portion.

[0072] Figs. 12A-G show views of a headband prototype according to the embodiment.

[0073] Figs. 13A-C show views of the headband prototype covered in a cover.

[0074] Fig. 13D shows an alternative headband prototype covered in a cover.DETAILED DESCRIPTION1. Overview

[0075] The embodiments described (for illustrative purposes only) broadly relate to a method and device for delivering stimulation to a user during a sleep period. It is desirable that the stimulation provided to the user assists with improving the restorative function of sleep without arousing the user from sleep. In this way, the stimulation provides the user with better sleep quality over a sleep period or session without increasing the time duration of the sleep period.

[0076] One or more embodiments described herein deliver stimulation tailored for the user. That is, rather than provide stimulation at a static time interval, audio frequency, duration, volume, or other stimulation parameter, the stimulation provided to the user may vary in time interval, audio frequency, duration, volume, or other stimulation parameter so as to increase the electrical activity in the user’s brain without arousing the user from sleep. By varying one or more stimulation parameters, the stimulation tailored to the user affects the sleep activity of the user (including increasing the brain activity of the user for example) without arousing the user from sleep so that the user is better rested at the end of the sleep period without having to increase the length of sleep time.2. General embodiments

[0077] Description turns to a general method and device embodiments for delivering stimulation to a user during a sleep period, which will be described with reference to Figs. 1 and 2.

[0078] Fig. 1 shows a general method embodiment 10. The method 10 delivers one or more stimulation responses (interchangeable with “stimulus” herein) that can be tailored to the user as follows. The method 10 begins at step 12 where one or more signals are received that are indicative of a user’s sleep activity over a sleep period. These one or more signals may be an EEG (electroencephalogram) signal but may also be obtained through other means including for example one or more signals indicative of the user’s: motion, heart activity, respiratory activity (i..e. one or more respiratory events), and other. Some signals may be received from a sensor, but this is not necessary and any received signal may be obtained through other means.

[0079] At step 14, the method 10 extracts one or more features of each of the one or more signals received at step 12. Examples of extracted received EEG signals may include for example: delta power, slow-wave counts, spectral activity (which could be in any one or more of the alpha, theta, and sigma bands), alpha intrusion, peak-to-peak amplitude, changes in spectral power over time, and the like. Examples of extracted features of signals indicative of the user’s motion may include for example: accelerometer-based movement detection, posture shift, one or more stillness patterns, and the like. Examples of extracted features of signals indicative of the user’s heart activity may include for example: heart rate, heart rate variability, respiration rate, oxygen saturation, and the like.

[0080] At step 16, the method 10 determines a stimulation response. Determination of the stimulation response includes determining whether to vary one or more stimulation response parameters. Examples of stimulation response parameters which may be varied include: sound volume, sound frequency, timing (for example timing relative to the phase angle of a sleep slow-wave), time duration of the stimulation response delivery, or the like. In one example, the stimulation response parameters being varied are or include volume and sound frequency. In one example, the stimulation response parameters being varied are or include timing relative to a phase angle of a slow-wave, volume, and sound frequency. The determination for whether to vary one or more stimulation response parameters is based on the extracted one or more features of the one or more signals at step 14.

[0081] At step 18, the method 10 delivers a stimulation response to the user based on the determined stimulation response at step 16. If it is determined at step 16 that it is desirable to vary one or more stimulation parameters, the stimulation delivery at step 18 may includevarying the stimulation parameter accordingly during delivery to the user. The stimulation delivered to the user may be auditory, but may additionally or alternatively be visual, haptic, or another type of sensory. In some embodiments, one or more of the signals received at step 12 may be filtered at step 19 so that the delivery of the stimulation response to the user at step 18 may be timed.

[0082] In some embodiments, the method 10 may also include using the user’s response to the delivered stimulation as feedback to alter future stimulation responses delivered to the patient.

[0083] For example, in some embodiments, the method 10 also includes sensing the user’s response to the delivered stimulation at step 20. In some embodiments, the one or more signals received at step 12 may be obtained from the sensed response of the user at step 20. In some embodiments, the step 20 of sensing the user’s response to the delivered stimulation includes sensing for increased electrical activity in the user’s brain. For example, the step 20 of sensing the user’s response to the delivered stimulation may include sensing for increased slow-wave activity (e.g. increased slow-wave count, increased delta power of an individual slow-wave) in the user’s brain.

[0084] In some embodiments, the method 10 also includes at step 22 the step of evaluating the user’s response to the stimulation response delivered at step 18. The evaluation at step 22 may be based directly on the sensed response of the user at step 20. Additionally or alternatively, the evaluation at step 22 may be based on the extracted features performed at step 14. Optionally, the evaluation may be performed by assigning a reward score for each delivered stimulation. The reward score may be indicative of how effective the delivered stimulation response was on improving the user’s sleep (which will be explained later). In some embodiments, the method may also include at step 24 updating a model based on the user’s evaluated response to the delivered stimulation at step 22. One or more outputs of the updated model may be used to modify how the determination at step 16 is performed.

[0085] In some embodiments, the method may also include suppressing stimulation if any received data suggests that any further stimulation delivery could arouse the user from sleep (not shown in Fig. 1).

[0086] In some embodiments, one or more steps of the method 10 may be performed by a device. For example, in Fig. 1, steps 12, 14, 16, 18, 19, and 20 are performed by device 30.

[0087] In some embodiments, one or more steps of the method 10 may be performed by a device and a server. For example, in Fig. 1, steps 12, 14, 16, 18, 19, and 20 are performed by device 30 while steps 22 and 24 are performed by a server 60. In the example of Fig. 1, The device 30 may communicate with a server 60 which may be in a different location to where the device 30 is.

[0088] In some embodiments, one or more controllers may be configured (for example by an algorithm) to perform one or more steps of the method 10.

[0089] Fig. 2 shows a general system embodiment 70. System 70 includes a general device embodiment 30 and optionally a general server embodiment 60. The general device embodiment 30 and / or general system embodiment 70 may be used to perform the method 10 as explained with reference to Fig. 1.

[0090] The device 30 includes a sensor 32, a controller 34, and an actuator 36. Although the example of Fig. 2 shows a single sensor, a single controller, and a single actuator, some device embodiments may have two or more sensors, controllers, and / or actuators.

[0091] The sensor 32 may be used to sense data relating to the user’s sleep. In some embodiments, the sensor 32 may be used to receive one or more signals indicative of the user’s sleep activity (step 12). In some embodiments, the sensor 32 may be used to sense the user’s response to the delivered stimulation (step 20). Examples of the sensor 32 may include EEG electrodes, heart rate sensor, motion sensor, respiratory monitoring, and the like.

[0092] The controller 34 may include a measurement module 42 to receive one or more signals (from the sensor 32 or otherwise) for the purpose of step 12. The controller 34 may include a feature extraction module 44 to extract one or more features (step 14) from the received signals (step 12). The controller 34 may include a stimulation determination module 46 to determine (step 16) whether to respond to the received signals with stimulation and if so, whether to vary any stimulation parameters in the stimulation response. The controller 34 may include a stimulation delivery module 48 to generate one or more signals to transmit to the actuator in order to the stimulation response at step 18. The controller 34 may includea Kalman filter 49 to filter (step 19) one or more received signals (step 12) to assist with the timing the delivery of the stimulation response to the user.

[0093] Although the example of Fig. 2 shows the measurement module 42, feature extraction module, stimulation determination module 46, stimulation delivery module 48, and Kalman filter 49 as part of the controller 34, any one or more of these components may exist outside of the device controller 34.

[0094] The actuator may be used to deliver stimulation to the user at step 18. Examples of the actuator 36 may include speaker, light, haptics, and the like.

[0095] The device 30 is communicable with an external server 60. The server 60 may include a response evaluation module 62 to evaluate the user’s monitored response to the delivered stimulation (step 22). The server 60 may also include a policy update module 64 to update a model (step 24) whose output can be used to modify how the determination at step 16 is performed. In one example, the server 60 is a centralised cloud-based service. The cloud-based service may be communicable with one or a plurality of devices 30.

[0096] In some embodiments, the device 30 is a wearable device. The wearable device can be a headband (such the headband shown in Figs. 8A-13C) for example.

[0097] Next, the description turns to exemplary embodiments falling within the scope of method 10, device 30, and system 70.3. Exemplary embodiments3.1. First exemplary embodiment

[0098] Description turns to a first exemplary embodiment with respect to Fig. 3.

[0099] Fig. 3 shows an exemplary method embodiment 110. The method 110 includes steps 112, 114, 116, 118, 120, 122, and 124 which will be described in more detail later. In the example of Fig. 3, the method 110 is performed using the exemplary system embodiment 170 which includes the exemplary device embodiment 130 and backend server 160.

[0100] The device 130 includes at least one sensor 132, at least one actuator 136, a feature extraction module 144, a policy engine 146, a stimulation delivery module 148, and a Kalman filter 149. The backend server 160 includes a response evaluation module 162 anda policy update module 164. In the example of Fig. 3, the device 130 is in the same environment 180 as the user 182 so that the one or more actuators 136 can deliver stimulation 118 to the user, and the one or more sensors 120 can sense the user’s 182 response to the delivered stimulation.

[0101] Description turns to how exemplary method 110 is performed with system 170 in detail.

[0102] At step 112, the feature extraction module 144 receives EEG, heart rate (HR) and motion signals indicative of the user’s sleep activity during the sleeping period. These signals can be acquired from one or more sensors 132 (worn by the user 182) including EEG, heart rate, and motion sensors. Additionally or alternatively, the signals can be acquired through other means.

[0103] At step 114, the feature extraction module 144 extracts the features including the user’s sleep stage / state, and spectral features. The feature extraction module 144 runs continuously on the device 130 and processes biosignals to generate features that support stimulation timing, sleep staging, policy decisions, and long-term analysis. The feature extraction module 144 operates on data from EEG, heart rate, and motion sensors 132. The features extracted by the feature extraction module 144 may include one or more of the following from one or more of the following types of signals:The feature extraction module 144 may also extract one or more features of any other biosignal relevant to sleep state or arousal detection. The feature extraction module 144 computes these features using signal processing tools such as for example filtering, spectral analysis (eg. FFT), rolling averages, and pattern detection. In some embodiments, Al modelling or other machine learning methods may be used. The signal processing tools provide a flexible data layer to support any system function that depends on understanding user state.

[0104] At step 116, the policy engine 146 determines from the extracted features a stimulation response. The policy engine 146 whether stimulation should be delivered to the user 182. If so, the policy engine 146 considers whether to vary any of the timing (relative to the phase angle of the slow-wave), sound frequency, volume, and stimulation duration when delivering the stimulation to the user 182. The policy engine 146 runs on a combination of global rules and user-specific adjustments. The policy engine 146 operates on fixed logic that applies universally across user 182 and on other users on other devices 130, such as avoiding stimulation during light sleep or periods of potential arousal or waking. Optionally, the policy engine 146 may also incorporate personalised decision-making that evolves based on how an individual responds over time. In one example, the policy engine 146 might reduce volume after signs of user arousal, shift timing based on sleep depth, or skip stimulation if delta power is already high or recent stimulation responses were ineffective. The policy engine 146 outputs a specific set of stimulation parameter values, which are transmitted to the stimulation delivery module 148.

[0105] At step 118, the stimulation delivery module 148 generates and transmits the stimulation signal to one or more actuators 136 in order to deliver stimulation to the user 182. In the example of Fig. 3, the actuator 136 is a bone-conduction speaker.

[0106] The stimulation delivery is timed by the Kalman filter 149 at step 119. The provision of the Kalman filter 149 enables precise timing of the stimulation delivery. The Kalman fdter 149 estimates the phase of the user’s slow-wave activity in real time. The Kalman filter 149 operates on incoming EEG data (at step 112) and models the underlying oscillatory signal as a hidden state that evolves over time:• The Kalman filter 149 models an EEG signal as a sinusoidal oscillation: x(t) = A cos(cot + (p).• The Kalman filter 149 tracks this signal’s phase (cp) and indirectly the frequency (co) over time, even in the presence of noise.• The Kalman filter 149 represents the signal in a 2D state space: the cosine and sine components of the signal, which evolve as a rotation in the complex plane.At each step, the filter predicts the next phase and then corrects based on observed EEG values. This allows the ability to estimate the current phase with reduced sensitivity to noise and artifacts. Optionally, the Kalman filter 149 may incorporate frequency variability. Optionally, the Kalman filter 149 is an extended Kalman filter.

[0107] The exemplary method 110 and system 170 also makes use of the user’s response to the delivered stimulation as feedback to refine future delivery of stimuli to the user 182. That is, the method 110 and system 170 can adjust one or more stimulation parameters (i.e. timing, frequency, volume, and duration) based on the user’s measured neurological responses for delivering future stimulation to the user 182. This facilitates personalised optimisation of stimulation parameters based on the user’s individual brain activity. Such adaptation, refinement, or modification of stimulation parameters to deliver the desired future stimuli can take place over a single sleep session or over multiple sleep sessions to improve efficacy.

[0108] At step 120, the EEG, heart rate, and motion sensors 132 sense the user’s 182 response to the delivered stimulation. The signals induced in the sensors 132 are transmitted for the device 130 to use, thus repeating step 112.

[0109] The signals induced in the sensors 132 are also transmitted to the backend server 160.

[0110] At step 122, the response evaluation module 162 evaluates the user’s monitored response to the previously delivered stimulation. That is, the response evaluation module reviews the EEG signal in the seconds following the delivered stimulation to evaluate how the user’s brain responded. The evaluation is performed by linking the delivered stimulation response to the stimulation parameters applied at the time, i.e. phase timing, volume, frequency, and time duration. From the EEG, the response evaluation module 162 extracts features such as:• The amplitude and shape of the first, second, and third slow-wave peaks after the stimulus.• Change in delta power over a short window (for example a time period between 5 to 10 seconds) compared to baseline.• Presence or absence of signs of arousal, such as high-frequency bursts.The extracted features are used to calculate a response score for each delivered stimulation. Over time, the response evaluation module 162 generates a profile of which stimulation parameter combinations lead to the most effective and least disruptive responses (in user 182 and other users), and adapts future stimulations accordingly.

[0111] At step 124, the policy update module 164 updates a model or ruleset to improve future determinations by the policy engine 146. This may include one or more of: updating weights in a scoring function, adjusting thresholds, or storing successful stimulation parameter combinations to reuse.

[0112] Thus, the exemplary system 170 is a closed loop system that adapts the stimulation parameters dynamically in response to ongoing and historical measures of the user’s brain response, rather than rely on fixed settings (i.e. fixed stimulation parameters) when delivering stimulation to the user.3.1.1. Variations

[0113] Variations of the exemplary method 110 and 170 will now be described.

[0114] Although the exemplary method 110 and system 170 uses sound as stimulation, other forms of stimulation may additionally or alternatively be provided. This may include for example: light, haptic (vibration), or magnetic stimulation, with appropriate or analogous substitution of control parameters (e.g., brightness, vibration intensity, magnetic field strength, power level).

[0115] Although the exemplary method 110 and system 170 use EEG signals to monitor or measure the user’s brain activity in real-time, other real-time brain monitoring technologies can alternatively or additionally be used to guide stimulation. This may include one or more of: fNIRS, fMRI, or emerging biosignal modalities.

[0116] The phase estimation by the Kalman filter may be substituted with any algorithm capable of accurately estimating phase in real time. This may include a variation of Kalman filtering (e.g. extended Kalman filter), or future predictive models.

[0117] Stimulation parameters (including timing, volume, frequency, and duration) can be adapted in real time or over multiple sleep sessions (over multiple nights for example) based on historical brain activity data, ongoing user state, or environmental conditions.

[0118] The specific parameters being adapted — volume, sound frequency, timing, and duration — may be expanded or replaced with modality -specific equivalents depending on the stimulus used.

[0119] Optionally, a calibration or onboarding stage / phase may be provided to collect baseline data and refine the stimulation parameters to adapt. That is, a calibration or onboarding phase may be included to collect user-specific data but is not required.

[0120] Optionally, the system 170 may begin adapting immediately using general heuristics. That is, the system 170 can begin adapting based on generalised defaults refined through ongoing use.

[0121] Although method 110 and system 170 stimulates delta wave or slow-waves in the user, it will be apparent that other brain activity may be targeted.

[0122] Additional variations may include one or more of the following:• Longitudinal adaptation: Stimulation parameters may be adjusted not only in real time but also over multiple nights, using historical response data to improve efficacy, reduce arousals, or account for changing sleep patterns.• Selective suppression: The system may choose to withhold stimulation when delta power is already low, or when EEG, movement, or biometric signals suggest increased arousal risk, to avoid overstimulation.• Multimodal input: Additional inputs such as heart rate variability, respiratory patterns, ambient noise levels, or circadian timing may be used to further tailor stimulation parameters.• Extended targeting: While the current implementation targets delta waves, the system may be used to modulate other brain states (e.g. alpha suppression, theta enhancement) during sleep or wake.• Stimulation sequencing: The system may apply sequences or clusters of stimulation cues rather than single pulses, with timing patterns optimized for extended slow- wave reinforcement.• Population-based learning: Algorithms may group users into response profiles or clusters to personalize parameter selection based on expected responsiveness.• Other modalities: In addition to sound, light, haptics, or magnetic stimulation, the system may employ emerging stimulation methods such as low-intensity focused ultrasound (FUS), with parameters including pulse duration, focus depth, and intensity adapted analogously.3.2. Second exemplary embodiment

[0123] Description turns to a further exemplary embodiment (not shown in figures). In this exemplary embodiment, it is the sound frequency, volume, and timing that is varied.

[0124] This exemplary embodiment relates to a multi-factorial algorithm for phase- targeted auditory stimulation, designed to create a personalised response for each user. The algorithm optimises, varies, or adjusts the sound frequency, volume, and timing of auditory stimuli based on individual user characteristics, providing efficient power consumption and enhanced efficacy. The algorithm dynamically adjusts, timing, volume, and sound frequency in real-time to achieve the desired neurological response from the user.

[0125] Personalised Sound Frequency and Volume Adjustment:• This exemplary embodiment uses a personalized approach to adjust the sound frequency and volume of auditory stimuli.• The algorithm measures the user’s neurological response to various combinations of sound frequencies and volumes to identify a suitable response for the user.• This personalised adjustment enhances the efficacy of stimulation and improves power efficiency by balancing frequency and volume.

[0126] Adaptive timing adjustment:• Traditional algorithms trigger auditory stimulation at a fixed point in the slow-wave oscillation.• This exemplary embodiment uses a dynamic timing mechanism that adjusts the trigger point based on the user’s neurological response.• To account for differences in neurological responses among users, such as elderly individuals with slower response times for example, the algorithm may stimulate earlier or later in a slow-wave oscillation to deliver a suitable response for the user.• The timing of stimulation delivery is continuously monitored and adjusted to find the optimal response point for each user, providing personalised efficacy.

[0127] This multi-factorial algorithm provides a personalised approach to phase-targeted auditory stimulation, dynamically adjusting volume, sound frequency, and timing to improve user response and improve power efficacy. The integration of real-time monitoring and continuous adjustment allows the embodiment to adapt to changing conditions and user profiles, providing an effective and efficient solution for improving sleep quality.3.3. Third exemplary embodiment

[0128] Description turns to a further exemplary embodiment (not shown in figures). In this exemplary embodiment, it is the sound frequency, volume, and timing that is varied.

[0129] This exemplary embodiment relates to a multi-factorial algorithm for phase- targeted auditory stimulation, designed to create a personalised response for each user. The algorithm optimises, varies, or adjusts the sound frequency, volume, and timing of auditory stimuli based on individual user characteristics, providing efficient power consumption and enhanced efficacy. The algorithm dynamically adjusts, timing, volume, and sound frequency in real-time to achieve the desired neurological response from the user.

[0130] Personalised Sound Frequency and Volume Adjustment:• The algorithm assesses the user’s neurological response to a range of sound frequencies and volumes. This involves monitoring neurological responses to various tones at different frequencies and volumes.• Based on the collected data, the algorithm identifies the combination of sound frequency and volume where the user exhibits the greatest response (which will be explained in more detail later) and adjusts the auditory stimuli accordingly to the identified frequency and volume ranges.• The algorithm continues to adjust these parameters in real-time, using historical user responses to refine the optimal combination for each user.

[0131] Adaptive Timing Mechanism:• The algorithm monitors the user’s slow- wave oscillations and identifies a suitable trigger point for auditory stimuli.• To account for differences in neurological responses among users, the timing is adjusted to match the user’s response, stimulating earlier or later in the slow-wave oscillation as needed.• The algorithm continuously refines the timing based on real-time feedback and historical user responses to ensure the best response.

[0132] Real-Time Volume and Sound Frequency Control:• The algorithm employs a dynamic control system to adjust volume and sound frequency in real-time.• It continuously monitors the user’ s neurological responses and adjusts the parameters to maintain optimal conditions.• The algorithm uses historical data to inform adjustments, ensuring that the combination of volume and sound frequency remains effective as user conditions change.

[0133] This multi-factorial algorithm provides a personalised approach to phase-targeted auditory stimulation, dynamically adjusting volume, sound frequency, and timing to improve user response and improve power efficacy. The integration of real-time monitoring and continuous adjustment allows the embodiment to adapt to changing conditions and user profiles, providing an effective and efficient solution for improving sleep quality.4. User response to the delivered stimulation

[0134] The description now explains how the delivered stimulation impacts the brain activity of the user to provide better sleep for the user without needing to increasing the duration of the sleep period. To demonstrate this, the description turns to simulations of the user delivered stimulation by way of reference with respect to Figs. 4-7.

[0135] Fig. 4 shows an ERP (Event-Related Potential) response graph of a stimulation delivery method and system (e.g. method 110 and system 170). The x-axis is time (in seconds), and the y-axis is voltage (in microvolts pV). Fig. 4 includes a STIM waveform 202 which shows the user’s electrical activity in the brain as it responds to the delivered stimulation over time. The delivered stimuli are shown in Fig. 4 as dots at 204a-e. The five dots of 204a-e represent five rounds of stimulation delivered to the user. In Fig. 4, the dots 204a-e near the peaks represent the point in time that each of the five stimulation delivery rounds is applied, and have no reference in the y-axis. Fig. 4 also shows a SHAM waveform 206 which provides a placebo reference. That is, in the example of Fig. 4, the SHAM waveform 206 represents an absence of delivered stimulation to the user.

[0136] In the example of Fig. 4, the volume of the delivered stimulation was adjusted. Pink noise frequency was selected for delivering the stimulation, and phase was a measured static degree from slope.

[0137] In the example of Fig. 4, the system delivers stimuli to the user on a five-on / five- off protocol (for example throughout the sleeping period), which allows comparison between a stimulation train and a non- stimulation train such that the STIM and SHAM waveform are can be superimposed on each other in Fig. 4. The average STIM and SHAM event throughout the sleeping period are averaged to plot the ERP of Fig. 4. The five rounds of stimulation delivered by 204a-e show an increase in electrical activity (from the increased amplitude of the voltage waveform), as well as an increase in slow-waves (induced by the stimulation) in the STIM waveform 202 (compared to the SHAM waveform 206). The increase in electrical activity and slow-wave count as shown by the STIM waveform 202 indicate that the stimulation delivered to the user has achieved its desired effect of providing better sleep for the user without needing to increasing the duration of the sleep period.

[0138] Fig. 5A shows another ERP (Event-Related Potential) response graph of a stimulation delivery method and system (e.g. method 110 and system 170). The x-axis is time (in seconds), and the y-axis is voltage (in microvolts pV). Fig. 5A includes a STIM waveform 212a which shows the user’s electrical activity in the brain as it responds to the delivered stimulation over time. The delivered stimuli are shown in Fig. 5 A as dots at 214a- d. The four dots of 214a-d represent five rounds of stimulation delivered to the user. In Fig. 5A, the dots 214a-d near the peaks represent the point in time that each of the four stimulation delivery rounds is applied, and have no reference in the y-axis. Fig. 5A also shows a SHAM waveform 216a which provides a control reference. In the example of Fig. 5A, the SHAM waveform 206 represents four rounds of stimulation delivered to the user using a prior art system as depicted by the four crosses 218a-d. The crosses 218a-d represent the point in time that each of the four stimulation delivery rounds is applied by the control delivery system, and have no reference in the y-axis.

[0139] In the example of Fig. 5A, the system delivers stimuli to the user on a four-on / four- off protocol (for example throughout the sleeping period), which allows comparison between a stimulation train and a non- stimulation train such that the STIM and SHAM waveform are can be superimposed on each other in Fig. 5A. The average STIM and SHAM event throughout the sleeping period are averaged to plot the ERP of Fig. 5A. The four rounds of stimulation delivered by 214a-d show an increase in electrical activity (from the increased amplitude of the voltage waveform), as well as an increase in slow-waves (induced by the stimulation) in the STIM waveform 212a (compared to the SHAM waveform 216a).

[0140] The increased electrical activity and increased slow-wave count in the STIM waveform 212a (compared to the SHAM waveform 216a) is brought about by the difference in the timing of the delivered stimulation of the present disclosure (e.g. method 110 and system 170) compared to the existing prior art stimulation delivery system (i.e. SHAM waveform 216a). In the example of Fig. 5A, the method and system of the present disclosure (e.g. method 110 and 170) varies timing of when the stimulation 214a-d is delivered to the user, while in contrast the prior art system timing for the stimulation delivery 218a-d is SHAM:• Both the present disclosure (e.g. method 110 and system 170) and prior art system deliver the first round of stimulation at the same time (relative to phase angle), asshown in Fig. 5A as dot 214a and cross 218b are plotted in the same position along the x-axis.• However, for the second, third, and fourth rounds of delivered stimulation, Fig. 5A shows that the timing of the delivered stimulation as shown by dots 214b-d deviates from crosses 218b-d along the x-axis. For the second delivered stimulation round, it is the present disclosure (e.g. method 110 and system 170) that delivers stimulation (dot 214b) to the user slightly before the prior art system does (cross 218b). For the third delivered stimulation round, it is the present disclosure (e.g. method 110 and system 170) that delivers stimulation (dot 214c) to the user slightly before the prior art system does (cross 218c). For the fourth delivered stimulation round, it is the present disclosure (e.g. method 110 and system 170) that delivers stimulation (dot 214d) to the user slightly before the prior art system does (cross 218d).

[0141] The increase in electrical activity achieved by the method and system of the present disclosure (e.g. method 110 and system 170) compared to the prior art is also evident in the polar graph of Fig. 5B, in which the maximum STIM amplitude 212b is greater than the maximum SHAM amplitude 216b.

[0142] In addition, the polar graph of Fig. 5B also shows the greatest response at -45 degrees. This shows the efficacy of the present disclosure to stimulation focusing on a variable phase angle.

[0143] The increase in electrical activity and slow-wave count as shown by the STIM waveform 212a compared to the SHAM waveform 216a indicate that the stimulation delivered to the user by the present disclosure (e.g. method 110 and system 170) has achieved its desired effect of providing better sleep for the user without needing to increasing the duration of the sleep period.

[0144] Fig. 6A shows another ERP (Event-Related Potential) response graph of a stimulation delivery system (e.g. method 110 and system 170). Fig. 6A discloses similar waveforms and stimulation protocol to the graph of Fig. 5A, and hence the description and interpretation of Fig. 6A will not be provided again with respect to these similarities for the purpose of brevity.

[0145] Fig. 6A differs from Fig. 5A in that there is a second STIM waveform 222 on the graph. The second STIM waveform 222 is induced by two rounds of stimulation delivered by the method and system of the present disclosure (e.g. method 110, system 170). The two rounds of stimulation is depicted as two horizontal lines in Fig. 6A, with the first stimulation depicted as line 224a, and the second stimulation depicted as line 224b.

[0146] In the example of Fig. 6A, the present disclosure (e.g. method 110 and system 170) varies the delivered stimulation by varying timing in the position along the wave, and varying the time duration of the stimulation (in contrast existing techniques typically time the delivery 20 to 30 degrees from the peak of a slow wave and apply a time duration of 50ms). The STIM waveform 222 shows a shallow slow-wave. This is the type of brain activity that may be seen in an older individual, or in other individuals who naturally have a shallower and longer slow-wave activity. For the first slow-wave, the first stimulation 224a begins at around -70 degrees from the peak, with the stimulation delivery lasting for 200 milliseconds. For the second slow-wave, the second stimulation 224b begins around -50 degrees from the peak, with the stimulation delivery time lasting for 100 milliseconds. Both stimulation rounds 224a-b are considered early as the delivery first begins before the peak of their respective slow-wave.

[0147] Fig. 6B shows another ERP (Event-Related Potential) response graph. Fig. 6B discloses similar waveforms and stimulation protocol to the graph of Fig. 5A, and hence the description and interpretation of Fig. 6B will not be provided again with respect to these similarities for the purpose of brevity.

[0148] Similar to Fig. 6A, Fig. 6B differs from Fig. 5A in that there is a second STIM waveform 232 on the graph. The second STIM waveform 232 is induced by a single round of stimulation delivered. The single stimulation round is depicted as a single dot 234 in Fig. 6B.

[0149] The purpose of the example of Fig. 6B is to demonstrate the possible adverse effect on the slow-wave if stimulation is applied too early. The single stimulation round 234 is delivered to the patient at -90 degrees from the peak, or below the zero point. The relatively early timing of the stimulation delivery undesirably decreases slow-wave activity and delta power.

[0150] Fig. 6C shows another ERP (Event-Related Potential) response graph. Fig. 6C discloses similar waveforms and stimulation protocol to the graph of Fig. 5A, and hence the description and interpretation of Fig. 6C will not be provided again with respect to these similarities for the purpose of brevity.

[0151] Similar to Fig. 6A-B, Fig. 6C differs from Fig. 5 A in that there is a second STIM waveform 242 on the graph. The second STIM waveform 242 is induced by a single round of stimulation delivered. The single stimulation round is depicted as a single line 244 in Fig. 6C.

[0152] The purpose of the example of Fig. 6C is to demonstrate the possible adverse effect on the slow-wave if stimulation is applied for too long. As can be seen in Fig. 6C, the slow- wave activity and delta activity are dampened.

[0153] Fig. 7 shows a chart that displays the sound frequency to power component of the algorithm. Pink noise, the commonly used method, is used because it covers the entire sound spectrum, meaning a portion of that sound spectrum will be audible to the person, and will likely elicit the peak response. However, this has an excessive power draw which is not necessary. The diagram displays pink noise, and an individual's response rates within specific frequencies. In this example, it can be seen that power draw in the 250Hz band and 350Hz band elicits the greatest response (x represents peak response) for the lower power draw.5. Exemplary embodiment design

[0154] Description turns to the design of a headband device embodiment with respect to Figs. 8A-13D.

[0155] Although the term “device” as used in this section can refer to any of the earlier described embodiments (including device 30, device 130, and variations and combinations thereof), or in other words, be used to deliver stimulation to a user during a sleep session as described heretofore, it is apparent that this is not essential, and that the physical design as disclosed can be adopted for a different application (besides delivering stimulation during sleep).

[0156] The physical design described in this section may be applied to any device that incorporates any features and functions as disclosed in earlier sections. For example, thephysical design described in this section (including variations and combinations) may be applied to any device with one or more features and functions of the already described with respect to device 30, device 130, as well as any of their variations.

[0157] The headband embodiment described in this section has a first interface portion and a second interface portion. The interface portions interface with a head of the user. The headband embodiments have one or more sensors to sense one or more signals from the user. The headband embodiments have one or more actuators to deliver stimulation to the user. Operation of the one or more sensors and actuators are controlled by one or more controllers housed in the headband.

[0158] The headband embodiment is designed such that the first interface portion contains more electronics than the second interface portion. That is, most of the electronics in the headband are located in the first interface portion. Such design configuration provides for a more compact and sleeker look, because when the headband is in use, most of the electronics rest on the back side or portion of the user’s head, thus improving. In contrast, existing headbands have most of the electronics rest on the forehead or top of the user’s head during use typically resulting in a 15-25mm large hunk of plastic which sits on the user’s forehead or top of head, which is where most of the nerve endings are, such that the user can feel the headband weight. The size of existing designs also introduces interference with sleep surfaces (e.g. pillow, mattress, etc).5.1. Headband embodiment

[0159] Figs. 8A-1 IB depict a headband device embodiment 330.

[0160] Turning to Figs. 8A-B and 11A-B, the headband 330 includes a first interface portion 332 and a second interface portion 334. The headband 330 also includes a first flexible arm 336 and a second flexible arm 338. The second interface portion 334 is located on or along the first flexible arm 336.

[0161] A first end 340 of the first interface portion 332 connects (i.e. mechanically couples) to a first end 342 of the first flexible arm 336. A first clip portion 344a is located at a second end 346 of the first flexible arm 336. A second end 348 of the first interface portion 332 connects (i.e. mechanically couples) to a first end 350 of the second flexible arm 338. A second clip portion 344b is located at a second end 352 of the second flexible arm 338. When the headband 330 is in use, the first and second clip portions 344a-b clip together onthe front side or portion of the user’s head to fasten the headband 330 to a user (e.g. Fig. 12G).

[0162] An FPC (flexible printed circuit) 349 runs from the first interface portion 332 to the second interface portion 334 via the first flexible arm 336. The FPC 349 connects the electronics from the first interface portion 332 to the second interface portion 334. The first flexible arm 336 thus contains or controls the connection of the FPC 349 to the electronics of the second interface portion 334.

[0163] The first and second flexible arms 336, 338 contain relief points (not shown) which control the area of flex so that the headband folds up when not in use. The relief points also control the extent that the FPC 349 bends so that the FPC 349 does not over-flex.5.1.1. First interface portion

[0164] The first interface portion 332 will be described in detail with reference to Figs. 8C-D, 8L-N, 9A-B, and 11A.

[0165] Figs. 8C-D show close-up views of the first interface portion 332. The first interface portion 332 is a rear interface portion. That is, the first interface portion 332 is a portion of the headband that interfaces with the rear side or portion of the user’s head when the headband is in use (e.g. Figs. 12A-E).

[0166] Fig. 8C is a perspective view of a side of the first interface portion 332 that interfaces with the user in use. The first interface portion 332 has a first electrode 354, and a second electrode 356. The electrodes 354, 356 are sensors for sensing one or more signals from the user. The first electrode 354 has an op-amp to amplify any induced signal. The second electrode 356 does not have an op-amp to provide a reference.

[0167] The electrodes 354, 356 are each combs (i.e. each electrode has six comb members). The first and second electrodes 354, 356 are desirably shaped to make them easier to slide in and out of the fabric covering or sleeve, which will be described later. Although the example of Fig, 8C depict the first and second electrodes 356 as square shaped, the electrodes can alternatively be triangular in shape (see Fig. 13D).

[0168] The electrodes 354, 356 are removable. The electrodes 354, 356 are replaceable in the event that electrodes don't last the lifetime of the headband. The electrodes 354, 356 thus acts as a user-serviceable component, and enables the headband to be quickly and cost- effectively refurbished.

[0169] Fig. 8D is a perspective view of a side of the first interface portion 332 that faces away from the user in use. The first interface portion 332 has a first spring 358 that connects the first interface portion 332 to the first flexible arm 336. The first interface portion 332 has a second spring 360 that connects the first flexible arm 332 to the second flexible arm 338. The first and second springs 358, 360 allow for stretching between the first interface portion 332 and the flexible arms 336, 338 (e.g. Figs. 12C-E). The first and second springs 358, 360 are made of silicone. The first and second springs 358, 360 are removable and replaceable.

[0170] Silicone springs have a flat spring rate, so no matter how much stretch the headband is undergoing, it does not result in a "tighter" feeling. The default spring is quite light, as the headband doesn't need significant pressure. Too much pressure can introduce noise to the system. The headband comes with optional "heavier" spring rate silicone springs. That is, the first and second springs 358, 360 come in different tensions to allow the user to “tune” the sensation of tightness. However, the first and second springs 358, 360 do not directly tighten anything on the headband which prevents the headband from being over-tightened. This allows users to customise the subjective tightness of the headband. Though it maintains the same position on the head, no matter which spring weight is used.Referring to Figs. 8C-D, 8L-N, 9A-B, and 11 A, the first interface portion 332 includes four PCB’s (printed circuit boards) 351a-d. Four printed circuit boards 351a-d inwardly face the user when the headband is in use. The four printed circuit boards 35 la-d are contained within a respective housing 357a-d. The housings 357a-d are made of a flexible material, for example silicone.

[0171] The first interface portion 332 also includes two additional mechanical or physical control housings 359 a-b. Housings 359 a-b have the same look and are made of the same material as housings 357a-d. Housings 359a-b act as physical “stops” that prevent or reduce over-stretching or tearing of the FPC 349 as follows. Referring to Figs. 8L-N by way of example, a U-bend 353 in the FPC 349 is provided to manage stretching of the FPC 349. At Fig. 8M, the bend 353 sits inside a small cavity 355 within the physical control housing 359b. As the headband 330 stretches as shown in Fig. 8N, the bend 353 unfolds slightly, allowing movement without stressing the FPC 349. This keeps the headband thin, flexible, and durable, while maintaining a reliable electrical connection between the first and second interface portions 332, 334. In one example, the U-bend mechanism allows the FPC 349 to stretch while remaining only 3 -4mm thick. If the exemplary headband device embodiment330 is modified to optionally include an FPC inside mechanical or physical control housing 359a, this mechanical or physical housing can still provide the same physical “stop” function as mechanical or physical control housing 359b, despite not being shown in the figures.

[0172] Gaps 361a-e are provided between the housings 357a-d, 359a-b to create flex points between the printed circuit boards 351a-d. The flex points allow the electronics to conform to the curvature of the user’s head to provide comfort to the user (e.g. Fig. 12B).

[0173] The first interface portion 332 also includes a battery 363. The battery 363 is contained within a housing 365. The battery housing 365 also includes a charging coil (not shown) which enables the battery 363 to be wirelessly charged. The charging coil is a Qi coil which means any wireless phone charger can be used to charge the battery 363. One or more magnets (not shown) may be provided in the battery housing 365 to improve centering of the headband on the charging coil. The batter housing 365 is made of a flexible material, for example silicone.

[0174] The wireless charge coil means that it is not necessary to have any charge points. Having a wired connection for charging introduces electrical risk requiring mor electronic components and would make the overall headband design larger. Thus, it is preferable for the battery to not have any wired charging.

[0175] The battery 363 has a connection to the rest of the first interface portion 332 (and the rest of the headband 330) so that the battery can flex and slightly twist. The battery is the thickest part of the headband 330 at about 7mm. The battery 363 is a coin-cell which hangs off the bottom of the headband. When the headband is in use, the battery 363 is positioned below the apex of the occipital lobe (back of the head). This position and flex means the battery (which the thickest part of the headband), isn't noticeable as a component when the user is wearing the headband (e.g. Fig. 12B). In this way, the user does not feel battery 363 or battery housing 365, even when lying on a firm pillow.5.2.2. Second interface portion

[0176] The second interface portion 334 will be described in detail with reference to Figs. 8E, 10A-B, and 11A.

[0177] Fig. 8E show a close-up view of the second interface portion 334 and the clip portions 344a-b. In addition, a perspective view of the second interface portion 334 is also shown at Fig 10A and a side view of the second interface portion 334 is also shown at Fig. 10B. The second interface portion 334 is a front interface portion. That is, the secondinterface portion is a portion of the headband that interfaces with the front side or portion with the user’s head when the headband is in use (e.g. Figs. 12F-G).

[0178] With reference to Figs. 8E and 10A-B, and 11A the second interface portion 334 includes a first electrode 362, a second electrode 364, and a third electrode 366. The three electrodes 362, 364, 366 are sensors that sense one or more signals from the user. The second interface portion 334 also includes a heart rate monitor 368. The second interface portion 334 also includes a bone conduction speaker 370. For aesthetics, the bone conduction speaker 370 has a similar appearance to the three electrodes 362, 364, 366. The sensors 362, 364, 366, heart rate monitor 368, and bone conduction speaker 370 are electronically connected to the PCB’s 351a-d (in the first interface portion 332) via the FPC 349 (in the first flexible arm 336).

[0179] Electrodes 362, 364, 366 are housed in a removable component (not shown). In this way, the electrodes 362, 364, 366 come off as a single connected string, so that none of the independent electrodes pose a choking hazard. The string of electrodes is replaceable in the event that electrodes don't last the lifetime of the headband. The string of electrodes 362, 364, 366 thus acts as a user-serviceable component and enables the headband to be quickly and cost-effectively refurbished.

[0180] Electrode 362 has an op-amp which amplifies any induced signal. The op-amp sits within the ramped housing with the conductive electrode material at the top (not shown). There is a side-facing LED (not shown) within the housing of electrode 362 used to communicate device status with the user.

[0181] Electrode 364 acts as a reference, and as such does not have an op-amp. There is a side-facing LED (not shown) within the housing of electrode 364 used to communicate device status with the user.

[0182] Electrode 366 has an op-amp which amplifies any induced signal. The op-amp sits within the ramped housing with the conductive electrode material at the top (not shown). There is a side-facing LED (not shown) within the housing of electrode 366 used to communicate device status with the user.

[0183] Bone conduction speaker 370 appears as a mock version of electrodes 362, 364, 366. There is a side-facing LED (not shown) within the housing of bone conduction speaker 370 used to communicate device status with the user.5.2.3. Additional reference figures

[0184] Figs. 8F-K show top, bottom, and side views of the headband 330.

[0185] Figs. 12A-12G show a prototype of the headband 330 as described with respect to Figs. 8A-11B.

[0186] Figs. 8F-K and Figs. 12A-G show the same features that have already been described in detail with reference to Figs. 8A-E, 9A-B, 10A-B, and 11A-B. For the sake of brevity, these features are not described again with respect to Figs. 8F-K and Figs. 12A-G.5.2.4. Headband cover

[0187] Figs. 13A-C show the headband prototype 330 covered in a cover 372. The cover 372 is a sleeve. The cover 372 is made of fabric. The cover 372 is washable. The clip portions 344a-b can be unclipped so that the headband end containing the first flexible arm 336 and second interface portion 334 can slide in and out of the sleeved cover 372. The shape of electrodes 362, 364, 366, heart rate monitor 368, and bone conduction speaker 370 are shaped so that they can slide and be held by the sleeve cover 372. Although the cover 372 is shown as a sleeve in Figs. 13A-C, the cover 372 can be modified to also accommodate another component. For example, an eye mask can be built into the cover.

[0188] Fig. 13D shows an alternative headband prototype 330 covered in a cover 372. This alternative prototype has electrodes 354, 356 in a triangular shape, but otherwise has the same features and functions as the prototype depicted in Figs. 13A-C.6. Effects

[0189] One or more embodiments disclosed herein in may provide one or more of the following effects:• One or more embodiments of the disclosed method, system, and device enable precise real-time targeting of brain activity phase to enhance sleep through phase- locked stimulation.• One or more embodiments of the disclosed method, system, and device adapts or varies stimulation parameters — e.g. timing, sound frequency, volume, and duration — based on individual brain responses, improving efficacy across users.• One or more embodiments of the disclosed method, system, and device learns from historical brain activity data to continuously refine stimulation delivery performance over time.• One or more embodiments of the disclosed method, system, and device increases power efficiency by identifying and using only the most effective combinations of stimulation parameters, avoiding broad or unnecessary output.• One or more embodiments of the disclosed method, system, and device is compatible with multiple stimulation modalities and brain measurement techniques, allowing for future expansion or application beyond auditory systems.• One or more embodiments of the disclosed method, system, and device supports personalisation of stimulation delivery without the need for manual configuration by the user.• One or more embodiments of the disclosed headband maintains electrode contact without undue pressure / tightness and adapt to movement through the night.• One or more embodiments of the disclosed headband keeps a thin profile across the entire head. The thickest part is the battery at around 7mm, but its placement means it is not noticeable, and weight is balanced across the head, rather than being focused on one point like existing designs.• One or more embodiments of the disclosed headband includes an easily removable and washable cover.• The disclosed headband is flexible, including for example the flexibility of the electronics.7. Disclosed features

[0190] In view of the embodiments disclosed above, there is provided the following.

[0191] In one embodiment, there is provided a method 10, 110 for providing personalised stimulation to a user 182 during a sleep period. The method 10, 110 includes receiving one or more signals indicative of the user’s sleep activity during a sleep period 12, 112. Themethod 10, 110 includes extracting one or more features of the one or more signals 14, 114. The method 10, 110 includes determining a stimulation response including determining whether to vary one or more stimulation response parameters based on the extracted one or more features of the one or more signals 16, 116. The method 10, 110 includes delivering a stimulation response to the user based on the determined stimulation response including varying the stimulation response by varying one or more stimulation response parameters 18, 118.

[0192] In some embodiments, the delivered stimulation response improves the restorative function of the sleep period, optionally during deep sleep. In some embodiments, the method 10, 110 further includes the step of sensing the user’s response to the delivered stimulation 20, 120. In some embodiments, the step 12, 112 of receiving one or more signals indicative of the user’s sleep activity is based on the user’s sensed response 20, 120 to the delivered stimulation. In some embodiments, the method 10, 110 further includes the step of evaluating the user’s sensed response to the delivered stimulation 22, 122. In some embodiments, the step of evaluating the user’s sensed response to the delivered stimulation 22, 122 includes assigning a reward score for each delivered stimulation response. In some embodiments, the method 10, 110 further includes the step of updating a model based on the user’s evaluated response to the delivered stimulation 24, 124. In some embodiments, the step of determining a stimulation response 16, 116 is further based on one or more outputs of the updated model 24, 124. In some embodiments, the method 10, 110 further includes the step of filtering the one or more signals for timing the stimulation response delivery to the user 19, 119.

[0193] In some embodiments, the one or more stimulation response parameters includes one or more of: volume, sound frequency, timing, and time duration. In some embodiments, the one or more stimulation parameters are or includes volume and sound frequency. In some embodiments, the one or more stimulation parameters are or includes timing relative to a phase angle of a slow-wave, volume, and sound frequency. In some embodiments, the one or more signals includes one or more EEG signals of the user. In some embodiments, the one or more extracted features 14, 114 of the one or more EEG signals includes one or more of: delta power, slow-wave counts, spectral activity (in alpha, theta, and / or sigma bands for example), alpha intrusion, peak-to-peak amplitude, changes in spectral power over time,wave slope, and inter-peak duration. In some embodiments, the one or more signals includes one or more signals indicative of the user’s motion. In some embodiments, the one or more extracted features 14, 114 of the one or more signals indicative of the user’s motion includes one or more of: accelerometer-based movement detection, posture shift, and one or more stillness patterns. In some embodiments, the one or more signals includes one or more signals indicative of the user’s heart activity. In some embodiments, the one or more extracted features 14, 114 of the one or more signals indicative of the user’s heart activity includes one or more of: heart rate, heart rate variability, respiration rate, and oxygen saturation. In some embodiments, the one or more signals includes one or more signals indicative of the user’s one or more respiratory events. In some embodiments, the step of sensing the user’s response to the delivered stimulation 20, 120 includes sensing for increased electrical activity in the user’s brain. In some embodiments, the step of sensing the user’s response to the delivered stimulation 20, 120 includes sensing for increased slow- wave activity in the user’s brain prompted by the delivered stimulation.

[0194] In another embodiment, there is provided a device 30, 130, 330 for providing personalised stimulation to a user during a sleep period. The device 30, 130, 330 includes one or more controllers 34. The one or more controllers 34 are configured to receive one or more signals indicative of the user’s sleep activity during a sleep period 12, 112. The one or more controllers 34 are configured to extract one or more features of the one or more signals 14, 114. The one or more controllers 34 are configured to determine a stimulation response including determining whether to vary one or more stimulation response parameters based on the extracted one or more features of the one or more signals 16, 116. The one or more controllers 34 are configured to deliver a stimulation response to the user based on the determined stimulation response including varying the stimulation response by varying one or more stimulation response parameters 18, 118.

[0195] In some embodiments, the device 30, 130, 330 is user wearable. In some embodiments, the device 30, 130, 330 is a headband.

[0196] In some embodiments, the delivered stimulation response improves the restorative function of the sleep period, optionally during deep sleep. In some embodiments, the one or more controllers 34 are further configured to sense the user’s response to the delivered stimulation 20, 120. In some embodiments, receiving 12, 112 one or more signals indicativeof the user’s sleep activity is based on the user’s sensed response 20, 120 to the delivered stimulation. In some embodiments, the one or more controllers 34 are further configured to evaluate the user’s sensed response to the delivered stimulation 22, 122. In some embodiments, evaluating the user’s sensed response to the delivered stimulation 22, 122 includes assigning a reward score for each delivered stimulation response. In some embodiments, the one or more controllers 34 are further configured to update a model based on the user’s evaluated response to the delivered stimulation 24, 124. In some embodiments, determining a stimulation response 16, 116 is further based on one or more outputs of the updated model 24, 124. In some embodiments, the device 30, 130, 330 further includes a Kalman filter 49, 149 configured to filter the one or more signals for timing the stimulation response delivery to the user 19, 119.

[0197] In some embodiments, the one or more stimulation response parameters includes one or more of: volume, sound frequency, timing, and time duration. In some embodiments, the one or more stimulation parameters are or includes volume and sound frequency. In some embodiments, the one or more stimulation parameters are or includes timing relative to a phase angle of a slow-wave, volume, and sound frequency. In some embodiments, the one or more signals includes one or more EEG signals of the user. In some embodiments, the one or more extracted features 14, 114 of the one or more EEG signals includes one or more of: delta power, slow-wave counts, spectral activity (in alpha, theta, and / or sigma bands for example), alpha intrusion, peak-to-peak amplitude, changes in spectral power over time, wave slope, and inter-peak duration. In some embodiments, the one or more signals includes one or more signals indicative of the user’s motion. In some embodiments, the one or more extracted features 14, 114 of the one or more signals indicative of the user’s motion includes one or more of: accelerometer-based movement detection, posture shift, and one or more stillness patterns. In some embodiments, the one or more signals includes one or more signals indicative of the user’s heart activity. In some embodiments, the one or more extracted features 14, 114 of the one or more signals indicative of the user’s heart activity includes one or more of: heart rate, heart rate variability, respiration rate, and oxygen saturation. In some embodiments, the one or more signals includes one or more signals indicative of the user’s one or more respiratory events. In some embodiments, sensing the user’s response to the delivered stimulation includes monitoring for increased electricalactivity in the user’s brain. In some embodiments, sensing the user’s response to the delivered stimulation includes monitoring for increased slow-wave activity in the user’s brain prompted by the delivered stimulation.

[0198] In another embodiment, there is provided a headband device 330. The headband device 330 includes a first interface portion 332 to interface with a rear side of a user’s head. The headband device 330 includes a first flexible arm 336 physically coupled to the first interface portion 332. The headband device 330 includes a second interface portion 334 to interface with a front side of the user’s head. The second interface portion 334 located on the first flexible arm 336. The first flexible arm 334 includes a flexible printed circuit 349 or part thereof to electrically connect the first interface portion with the second interface portion.

[0199] In some embodiments, the first interface portion 332 is larger than the second interface portion 334. In some embodiments, the first interface portion 332 includes a battery 363. In some embodiments, the first and second interface portions 332, 334 are flexible. In some embodiments, the first interface portion 332 includes one or more printed circuit boards 351a-d. In some embodiments, each printed circuit board 351a-d is contained within a respective housing 357a-d. In some embodiments, the first interface portion 332 includes one or more physical control housings 359a-b. In some embodiments, gaps 361a-e are provided between the housings 357a-d, 359a-b to provide flex points. In some embodiments, the headband device 330 further includes a second flexible arm 338. In some embodiments, the headband device 330 further includes one or more flexible springs 358, 360 that connects between the first interface portion 332 and a respective flexible arm 336, 338. In some embodiments, the headband device 330 further includes a removable and / or washable cover 372.

[0200] In addition, the following clauses list a number of disclosed features as follows:1. A method for optimising auditory stimulation through personalised sound frequency and volume adjustments, including: measuring a user’s neurological response to a range of sound frequencies and volumes,adjusting the sound frequency and volume of auditory stimuli to the combination where the user has the greatest response, and enhancing the efficacy of stimulation while also optimising for power efficiency by balancing frequency and volume.2. A method for dynamically adjusting the timing of phase-targeted auditory stimulation based on individual neurological response, including: monitoring the user’s slow- wave oscillation, adjusting the auditory stimulus trigger point earlier or later in the oscillation to achieve the optimal stimulation response based on the user’s neurological response, and continuously optimising the trigger timing for each user based on real-time feedback.3. A method for real-time volume and sound frequency control in phase -targeted auditory stimulation, including: monitoring different volume levels at various sound frequencies to identify the most effective combination for the user, dynamically adjusting volume and sound frequency in response to the user’s neurological responses, and using historical user responses to inform real-time adjustments and optimise the desired characteristics of user response and battery performance.

[0201] Many modifications will be apparent to those skilled in the art without departing from the scope of the present invention.

[0202] The presence ofin a FIG. or text herein is understood to mean "and / or" unless otherwise indicated, i.e., “A / B” is understood to mean “A” or “B” or “A and B”. The recitation of a particular numerical value or value range herein is understood to include or be a recitation of an approximate numerical value or value range, for instance, within + / - 20%, + / - 15%, + / - 10%, + / - 5%, + / - 2.5%, + / - 2%, + / - 1%, + / - 0.5%, or + / - 0%. The term "essentially all" or "substantially" can indicate a percentage greater than or equal to 50%, 60%, 70%, 80%, or 90%, for instance, 92.5%, 95%, 97.5%, 99%, or 100%.

[0203] The reference in this specification to any prior publication (or information derived from it), or to any matter which is known, is not, and should not be taken as an acknowledgment or admission or any form of suggestion that the prior publication (or information derived from it) or known matter forms part of the common general knowledge in the field of endeavour to which this specification relates.

[0204] Throughout this specification and the claims which follow, unless the context requires otherwise, the word "comprise", and variations such as "comprises" and "comprising", will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

Claims

CLAIMS:

1. A method for providing personalised stimulation to a user during a sleep period comprising the steps of: receiving one or more signals indicative of the user’s sleep activity during a sleep period, extracting one or more features of the one or more signals, determining a stimulation response including determining whether to vary one or more stimulation response parameters based on the extracted one or more features of the one or more signals, and delivering a stimulation response to the user based on the determined stimulation response including varying the stimulation response by varying one or more stimulation response parameters.

2. A method for providing personalised stimulation according to claim 1, wherein the delivered stimulation response improves the restorative function of the sleep period, optionally during deep sleep.

3. A method for providing personalised stimulation according to claim 1 or 2, further comprising the step of sensing the user’s response to the delivered stimulation.

4. A method for providing personalised stimulation according to claim 3, wherein the step of receiving one or more signals indicative of the user’s sleep activity is based on the user’s sensed response to the delivered stimulation.

5. A method for providing personalised stimulation according to any one of the previous claims, further comprising the step of evaluating the user’s sensed response to the delivered stimulation.

6. A method for providing personalised stimulation according to claim 5, wherein the step of evaluating the user’s sensed response to the delivered stimulation includes assigning a reward score for each delivered stimulation response.

7. A method for providing personalised stimulation according to any one of the previous claims, further comprising the step of updating a model based on the user’s evaluated response to the delivered stimulation.

8. A method for providing personalised stimulation according to claim 7, wherein the step of determining a stimulation response is further based on one or more outputs of the updated model.

9. A method for providing personalised stimulation according to any one of the previous claims, further comprising the step of filtering the one or more signals for timing the stimulation response delivery to the user.

10. A method for providing personalised stimulation according to any one of the previous claims, wherein the one or more stimulation response parameters includes one or more of:• volume• sound frequency• timing• time duration11. A method for providing personalised stimulation according to any one of the previous claims, wherein the one or more stimulation parameters are or comprises volume and sound frequency.

12. A method for providing personalised stimulation according to any one of the previous claims, wherein the one or more stimulation parameters are or comprises timing relative to a phase angle of a slow-wave, volume, and sound frequency.

13. A method for providing personalised stimulation according to any one of the previous claims, wherein the one or more signals includes one or more EEG signals of the user.

14. A method for providing personalised stimulation according to claim 13, wherein the one or more extracted features of the one or more EEG signals includes one or more of:• delta power• slow-wave counts• spectral activity in alpha, theta, and / or sigma bands• alpha intrusion• peak-to-peak amplitude• changes in spectral power over time• wave slope• inter-peak duration.

15. A method for providing personalised stimulation according to any one of the previous claims, wherein the one or more signals includes one or more signals indicative of the user’s motion.

16. A method for providing personalised stimulation according to claim 15, wherein the one or more extracted features of the one or more signals indicative of the user’s motion includes one or more of:• accelerometer-based movement detection• posture shift• one or more stillness patterns17. A method for providing personalised stimulation according to any one of the previous claims, wherein the one or more signals includes one or more signals indicative of the user’s heart activity.

18. A method for providing personalised stimulation according to claim 17, wherein the one or more extracted features of the one or more signals indicative of the user’s heart activity includes one or more of:• heart rate• heart rate variability• respiration rate• oxygen saturation19. A method for providing personalised stimulation according to any one of the previous claims, wherein the one or more signals includes one or more signals indicative of the user’s one or more respiratory events.

20. A method for providing personalised stimulation according to any one of the previous claims, wherein the step of sensing the user’s response to the delivered stimulation includes sensing for increased electrical activity in the user’s brain.

21. A method for providing personalised stimulation according to any one of the previous claims, wherein the step of sensing the user’s response to the delivered stimulation includes sensing for increased slow-wave activity in the user’s brain prompted by the delivered stimulation.

22. A device for providing personalised stimulation to a user during a sleep period comprising one or more controllers configured to: receive one or more signals indicative of the user’s sleep activity during a sleep period, extract one or more features of the one or more signals, determine a stimulation response including determining whether to vary one or more stimulation response parameters based on the extracted one or more features of the one or more signals, and deliver a stimulation response to the user based on the determined stimulation response including varying the stimulation response by varying one or more stimulation response parameters.

23. A device for providing personalised stimulation according to claim 21, wherein the device is user wearable.

24. A device for providing personalised stimulation according to claim 22 or 23, wherein the device is a headband.

25. A device for providing personalised stimulation according to any one of claims 22 to24, wherein the delivered stimulation response improves the restorative function of the sleep period, optionally during deep sleep.

26. A device for providing personalised stimulation according to any one of claims 22 to25, wherein the one or more controllers are further configured to sense the user’s response to the delivered stimulation.

27. A device for providing personalised stimulation according to claim 26, wherein receiving one or more signals indicative of the user’s sleep activity is based on the user’s sensed response to the delivered stimulation.

28. A device for providing personalised stimulation according to any one of claims 22 to 27, wherein the one or more controllers are further configured to evaluate the user’s sensed response to the delivered stimulation.

29. A device for providing personalised stimulation according to claim 28, wherein evaluating the user’s sensed response to the delivered stimulation includes assigning a reward score for each delivered stimulation response.

30. A device for providing personalised stimulation according to any one of claims 22 to 29, wherein the one or more controllers are further configured to update a model based on the user’s evaluated response to the delivered stimulation.

31. A device for providing personalised stimulation according to claim 30, wherein determining a stimulation response is further based on one or more outputs of the updated model.

32. A device for providing personalised stimulation according to any one of claims 22 to31, further comprising a Kalman filter configured to filter the one or more signals for timing the stimulation response delivery to the user.

33. A device for providing personalised stimulation according to any one of claims 22 to32, wherein the one or more stimulation response parameters includes one or more of:• volume• sound frequency• timing• time duration34. A device for providing personalised stimulation according to any one of claims 22 to33, wherein the one or more stimulation parameters are or comprises volume and sound frequency.

35. A device for providing personalised stimulation according to any one of claims 22 to34, wherein the one or more stimulation parameters are or comprises timing relative to a phase angle of a slow-wave, volume, and sound frequency.

36. A device for providing personalised stimulation according to any one of claims 22 to35, wherein the one or more signals includes one or more EEG signals of the user.

37. A device for providing personalised stimulation according to claim 36, wherein the one or more extracted features of the one or more EEG signals includes one or more of:• delta power• slow-wave counts• spectral activity in alpha, theta, and / or sigma bands• alpha intrusion• peak-to-peak amplitude• changes in spectral power over time• wave slope• inter-peak duration.

38. A device for providing personalised stimulation according to any one of claims 22 to 37, wherein the one or more signals includes one or more signals indicative of the user’s motion.

39. A device for providing personalised stimulation according to claim 38, wherein the one or more extracted features of the one or more signals indicative of the user’s motion includes one or more of:• accelerometer-based movement detection• posture shift• one or more stillness patterns40. A device for providing personalised stimulation according to any one of claims 22 to 39, wherein the one or more signals includes one or more signals indicative of the user’s heart activity.

41. A device for providing personalised stimulation according to claim 40, wherein the one or more extracted features of the one or more signals indicative of the user’s heart activity includes one or more of:• heart rate• heart rate variability• respiration rate• oxygen saturation42. A device for providing personalised stimulation according to any one of claims 22 to 41, wherein the one or more signals includes one or more signals indicative of the user’s one or more respiratory events.

43. A device for providing personalised stimulation according to any one of claims 22 to42, wherein sensing the user’s response to the delivered stimulation includes monitoring for increased electrical activity in the user’s brain.

44. A device for providing personalised stimulation according to any one of claims 22 to43, wherein sensing the user’s response to the delivered stimulation includes monitoring for increased slow- wave activity in the user’s brain prompted by the delivered stimulation.

45. A headband device comprising: a first interface portion to interface with a rear side of a user’s head, a first flexible arm physically coupled to the first interface portion, a second interface portion to interface with a front side of the user’s head, the second interface portion located on the first flexible arm, and wherein the flexible arm includes a flexible printed circuit or part thereof to electrically connect the first interface portion with the second interface portion.

46. A headband device according to claim 45, wherein the first interface portion is larger than the second interface portion.

47. A headband device according to claim 45 or 46, wherein the first interface portion comprises a battery.

48. A headband device according to any one of claims 45 to 47, wherein the first and second interface portions are flexible.

49. A headband device according to any one of claims 45 to 48, wherein the first interface portion comprises one or more printed circuit boards.

50. A headband device according to claim 49, wherein each printed circuit board is contained within a respective housing.

51. A headband device according to claim 50, wherein the first interface portion comprises one or more physical control housings.

52. A headband device according to claim 50 or 51 , wherein gaps are provided between the housings to provide flex points.

53. A headband device according to any one of claims 45 to 52, further comprising a second flexible arm.

54. A headband device according to any one of claims 45 to 53, further comprising one or more flexible springs that connects between the first interface portion and a respective flexible arm.

55. A headband device according to any one of claims 45 to 54, further comprising a removable and / or washable cover.

Images