Wearable-based behavioral profiling for evaluating user morning alertness

Abstract

A method and system for determining morning alertness by leveraging the science of sleep inertia The method includes detecting sleep end time, monitoring heart rate and activity data post sleep end time. The method further includes calculating the active time based on the activity data of the user. Further, the method includes determining sleep inertia based on the sleep end time and the active time. The method further includes calculating a morning alertness score based on the sleep inertia. The morning alertness score is calculated by analysing the sleep inertia and activity data by a ML model. The morning alertness score offers insights into sleep effectiveness and psychological markers such as motivation.

Description

[0001] WEARABLE-BASED BEHAVIORAL PROFILING FOR EVALUATING USER MORNING ALERTNESS

[0002] FIELD OF INVENTION

[0003] [1] The present invention relates to a wearable device, and specifically relates to a method of assessing morning alertness of the user.

[0004] BACKGROUND

[0005] [2] The first 30-60 mins after waking up for a human being is a period of assessment from the previous day’s stressors and recovery that the previous night’s sleep session provided. In many cases, chronic strain on a person’s physiology and disease onset of peripheral and mental kind can show up as early signs that can be picked up by attuned, sensitive wearables. Morning alertness therefore becomes a metric critical to assess optimal daily functioning and is heavily influenced by the challenge of overcoming sleep inertia or other physiological challenges upon awakening. Sleep inertia is a physiological state of impaired cognitive and sensory-motor performance that is present immediately after awakening. These temporary refractory periods can persist for minutes to hours after waking up, significantly affecting a person's capacity to perform tasks effectively through the rest of the following active period of the day. Existing sleep tracking devices predominantly focus on quantifying sleep duration and quality, providing users with insights into their nightly rest patterns. The analysis ends when a person is detected to be awake. The period immediately after waking is hence not analyzed and behavior patterns there in, which can provide valuable insights into a user’s daily performance are missed. As a result, individuals may not receive accurate feedback on their readiness to engage in daily activities upon waking.

[0006] [3] With advances in medical research, morning alertness or lack thereof is being associated with early symptoms of conditions such as perimenopause, depression, ADHD and rare disorders such as narcolepsy. Hence quantifying this metric supports users in making decisions on not only their general wellbeing but also early disease onset detection.

[0007] [4] Thus, there is a need for a device system that can address this user gap and accurately assess the morning alertness of a user. OBJECTS OF THE INVENTION

[0008] [5] A general objective of the present invention is to provide a smart wearable capable of accurately determining the period immediately after waking of a user. This includes identifying sleep start and end through a combination of biomarkers that the wearable records via sensors.

[0009] [6] Another objective of the invention is to provide a smart wearable capable of determining a sleep inertia of the user.

[0010] [7] Yet another objective of the invention is to provide a smart wearable capable of assessing morning alertness of the user.

[0011] [8] Still another objective of the invention is to provide a smart wearable capable of using machine learning algorithms for assessing morning alertness of the user.

[0012] [9] Yet another objective of the invention is for the data for morning alertness stored in a decentralized server where historical patterns for a user can be stored and accessed for long term trend analysis.

[0013] SUMMARY OF THE INVENTION

[0014]

[0010] The summary is provided to introduce aspects related to a wearable device for assessing morning alertness of a user, and the aspects are further described below in the detailed description. This summary is not intended to identify essential features of the claimed subject matter nor is it intended for use in determining or limiting the scope of the claimed subject matter.

[0015]

[0011] The present invention relates to a wearable device for assessing morning alertness of a user. The wearable device may comprise a Photoplethysmography (PPG) sensor for collecting heart rate data of a user. The wearable device may further comprise a motion sensor for collecting activity data of the user. The wearable device may further comprise a sleep tracking sensor for determining a sleep end time of the user by analysing transitional states between different sleep stages of the user. The wearable device may further comprise a Central Processing Unit (CPU) communicatively coupled with a memory for calculating an active time of the user based on the sleep end time and the activity data. The active time corresponds to a significant increase in activity after the sleep end time. Further, the CPU determines an alertness score of the user based on the active time and the activity data, by analyzing quantity of activity performed by the user before the active time and after the sleep end time. Further, the CPU determines sleep inertia of the user by calculating a gap between the sleep end time and the active time.

[0016]

[0012] In one aspect, the sleep tracking sensor implements a sleep tracking algorithm for determining a wake-up time of the user.

[0017]

[0013] In one aspect, for determining the wake-up time of the user, the sleep tracking algorithm identifies a window of motion which could signify leaving the sleeping area and registering a predefined number of steps being taken within a predefined time period by the user.

[0018]

[0014] In one aspect, for determining the wake-up time of the user, the sleep tracking algorithm monitors the wearable device being placed for charging.

[0019]

[0015] In one aspect, the device and attendant computer application accesses historical data of habits and movement patterns of a user and cross verifies any deviations by user requested tagging.

[0020]

[0016] In one aspect, the motion sensor and the PPG sensor keep collecting the activity of the user post sleep end time.

[0021]

[0017] In one aspect, while determining the alertness score, the wearable device considers physiological parameters including heart rate, heart rate variability, respiration rate, and oxygen volume, temperature and motion.

[0022]

[0018] In one aspect, the gap indicative of the sleep inertia is quantified via a mathematical function, wherein the mathematical function is one of slope, numerical difference, and vector difference. Additionally, statistical analysis is applied to assess the likelihood of a pattern change in morning alertness. This is achieved by accumulating alertness-related deflection data across a predefined set of days (e.g., 5-7 days), and performing a comparative evaluation with the corresponding metrics from a preceding set of days. The comparison may utilize statistical techniques such as hypothesis testing, confidence intervals, or Bayesian inference to estimate the significance and likelihood of change, thereby enabling adaptive modeling of sleep inertia over time.

[0023]

[0019] In one aspect, the CPU further determines a morning alertness score of the user based on the sleep inertia, the heart rate, and the activity post active time.

[0020] The wearable device as claimed in claim 1, wherein the wearable device is one of a smart ring, smart watch, and smart band.

[0024]

[0021] In one aspect, the wearable device is connected with a user device such as smart phone, tablet, computer, or any IOT device for rendering the morning alertness score along with the activity data of the user.

[0025]

[0022] A method of determining morning alertness of a user is described. The method comprises collecting, using a PPG sensor, heart rate data of a user. The method further comprises collecting, using a motion sensor, activity data of the user. The method further comprises determining, using a sleep tracking sensor, a sleep end time of the user by analysing transitional states between different sleep stages of the user. The method further comprises calculating, by a Central Processing Unit (CPU), an active time of the user based on the sleep end time and the activity data. The active time corresponds to a significant increase in activity after the sleep end time. The method further comprises determining, by the CPU, an alertness score of the user based on the active time and the activity data, by analyzing quantity of activity performed by the user before the active time and after the sleep end time.

[0026]

[0023] In one aspect, the sleep tracking sensor implements a sleep tracking algorithm for determining a wake-up time of the user.

[0027]

[0024] In one aspect, the CPU removes raw data that is non-physiological. In other words, the CPU removes physiological outlier values for enhancing accuracy of the derived prediction.

[0028]

[0025] In one aspect, for determining the wake-up time of the user, the sleep tracking algorithm identifies the act of increased mobility by registering more than a predefined number of steps being taken within a predefined time period by the user.

[0029]

[0026] In one aspect, for determining the wake-up time of the user, the sleep tracking algorithm monitors the wearable device being placed for charging.

[0030]

[0027] In one aspect, the motion sensor and the PPG sensor keep collecting the activity of the user post sleep end time.

[0031]

[0028] In one aspect, while determining the alertness score, the wearable device considers physiological parameters including heart rate, respiration rate, and oxygen volume.

[0029] In one aspect, the gap indicative of the sleep inertia is quantified via a mathematical function, wherein the mathematical function is one of slope, numerical difference, and vector difference.

[0032]

[0030] In one aspect, the method further comprises determining sleep inertia of the user by calculating a gap between the sleep end time and the active time and determining a morning alertness score of the user based on the sleep inertia, the heart rate, and the activity post active time.

[0033]

[0031] In one aspect, the method is performed by a wearable device comprising a smart ring, smart watch, and smart band.

[0034]

[0032] In one aspect, the wearable device is connected with a user device for rendering the morning alertness score along with the activity data of the user.

[0035]

[0033] In one aspect, the wearable device communicates historical data of the user to a network cloud or a decentralized database for performing longitudinal trend analysis based on daily sleep inertia and morning alertness.

[0036] BRIEF DESCRIPTION OF THE DRAWINGS

[0037]

[0034] The accompanying drawings constitute a part of the description and are used to provide further understanding of the present invention. Such accompanying drawings illustrate the embodiments of the present invention which are used to describe the principles of the present invention. The embodiments are illustrated by way of example and not by way of limitation in the figures of the accompanying drawings in which like references indicate similar elements. It should be noted that references to “an” or “one” embodiment in this invention are not necessarily to the same embodiment, and they mean at least one. In the drawings:

[0038]

[0035] Fig. 1 illustrates a block diagram showing components of a wearable device for assessing morning alertness of a user, in accordance with an embodiment of the present invention;

[0039]

[0036] Fig. 2 illustrates a method of assessing morning alertness of a user, in accordance with an embodiment of the present invention; and

[0037] Fig. 3 illustrates a flow diagram of a method of assessing morning alertness of a user, in accordance with an embodiment of the present invention.

[0040] DETAILED DESCRIPTION OF THE INVENTION

[0041]

[0038] The detailed description set forth below in connection with the appended drawings is intended as a description of various embodiments of the present invention and is not intended to represent the only embodiments in which the present invention may be practiced. Each embodiment described in this disclosure is provided merely as an example or illustration of the present invention, and should not necessarily be construed as preferred or advantageous over other embodiments. The detailed description includes specific details for the purpose of providing a thorough understanding of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced without these specific details.

[0042]

[0039] The proposed invention relates to a system and method for assessing morning alertness of a user. The system may be a wearable device, such as smart watch, smart band, or an electronic ring. Although the details have been provided successively with reference to a smart ring merely for the sake of explanation, it must be understood that the invention could be fairly implemented in a similar manner using any other wearable device, such as the ones listed above.

[0043]

[0040] Fig. 1 illustrates a block diagram showing components of a wearable device 100 for assessing morning alertness of a user, in accordance with an embodiment of the present invention. In one implementation, the wearable device 100 may be a smart ring 100 (as illustrated in Fig. 1); however, it must be understood that present invention may be similarly implemented using other wearable devices like smart watch, smart band, etc. The smart ring 100 may be made using a hypoallergenic material for allowing comfortable and continuous wear by a user. The smart ring 100 may include hardware components and firmware components. The hardware components may comprise a Central Processing Unit (CPU) 102, a memory 104, a motion sensor 106, a sleep tracking sensor 108, a PPG sensor 110, a Bluetooth Low Energy (BLE) communication unit 112, and a battery and charging unit / circuitry 114. The motion sensor 106 may include an accelerometer. Further, the CPU may exist and operate alongside a Machine Learning (ML) co-processor 116 implemented as a software component within the memory 104. The firmware components may include a software database management module, the ML co-processor 116, and a sleep tracking algorithm. In some implementations, the ML co-processor 116 may be implemented over a cloud network with which the smart ring 100 may be present in communication.

[0044]

[0041] Fig. 2 illustrates a method of assessing morning alertness of a user, in accordance with an embodiment of the present invention. A method 200 of assessing morning alertness of a user is explained successively. At block 202, the smart ring 100 may determine a sleep end time of the user, utilizing the sleep tracking sensor 108. The sleep tracking sensor 108 may determine sleep patterns of the user using the sleep tracking algorithm. The sleep tracking algorithm may analyse transitional states between different sleep stages, ultimately pinpointing the sleep end time.

[0045]

[0042] The sleep tracking algorithm is an ML model trained to determine the user’s wake-up time. For determining when the user wakes up, the sleep tracking algorithm monitors the next four-hour window to confirm this event through step activity. Specifically, the sleep tracking algorithm checks whether the user takes three 50-80 steps in a predefined time (such as 15- minute) window each or shows a single instance of more than 100 steps. Alongside movement tracking, the sleep tracking algorithm also monitors if data is missing, if the smart ring 100 is placed on charge, or if there's a loss of contact indicating the smart ring 100 is not worn during this period. Although the sleep tracking algorithm doesn't calculate an exact point of cognitive alertness, it uses these signals to provide an informative wake-up insight to the user.

[0046]

[0043] At block 204, the smart ring 100 may collect heart rate data, utilizing the PPG sensor 110. The PPG sensor 110 may keep collecting the heart rate data of the user, post sleep end time. At block 206, the smart ring 100 may collect activity data of the user, utilizing the motion sensor 106. The motion sensor 106 may determine the activity of the user based on the acceleration and motion of the user determined by the accelerometer. The motion sensor 106 may keep collecting the activity data of the user post sleep end time.

[0047]

[0044] At block 208, the smart ring 100 may calculate an active time of the user based on the sleep end time and the activity data. The significant increase in the activity after the sleep end time may correspond to the active time of the user.

[0048]

[0045] At block 210, the smart ring 100 may determine an alertness score of the user based on the active time and the activity data. The smart ring 100 may determine the alertness score by analyzing the quantity of activity performed by the user before the active time and post sleep end time, along with other physiological parameters, such as heart rate, respiration rate, oxygen volume etc.

[0049]

[0046] At block 212, the smart ring 100 may determine sleep inertia of the user based on the sleep end time and the active / alert time. The smart ring 100 may determine the sleep inertia by calculating a gap between the active time and the sleep end time (alert time). The gap may be quantified via a mathematical operation or function, such as slope, numerical difference, or vector difference.

[0050]

[0047] At block 214, the smart ring 100 may determine a morning alertness score of the user based on the sleep inertia. The smart ring 100 may analyze the sleep inertia, the heart rate, and the activity post active time using a Machine Learning (ML) model to determine a morning alertness score of the user. The ML model may be trained using supervised ML techniques such as Support Vector Machines, Decision Trees, Random Forests, and Naive Bayes.

[0051]

[0048] Fig. 3 illustrates a flow diagram of a method 300 of assessing morning alertness of a user, in accordance with an embodiment of the present invention. The method 300 may be implemented by the CPU of the smart ring. The method 300 may include detecting sleep end time of the user, at step 302. The sleep end time is determined based on the sleep pattern of the user using the sleep tracking algorithm. The sleep tracking algorithm may analyse the transitional states between different sleep stages to calculate a sleep end time. In one implementation, the sleep tracking algorithm may determine the sleep end time by comparing the sleep pattern with a sleep profile of the user. The sleep profile may be generated when the user continuously wears the smart ring for several days.

[0052]

[0049] Further, the method 300 includes collecting the heart rate data and the activity data of the user, at step 304. The heart rate data is collected using the PPG sensor, post sleep end time. The activity data is collected using the motion sensor, post sleep end time. In one implementation, the PPG sensor and the motion sensor may start collecting activity data at a pre-defined interval, such as 1 min from 30 mins before sleep end time based on the sleep profile of the user.

[0053]

[0050] The method 300 further includes calculating an active time of the user post sleep end time, at step 306. The active time is determined when the smart ring detects a significant increase in the activity or physiological signatures of the user. The active data is a time stamp of the incident of the significant increase in activity of the user. In an embodiment, the active time may be calculated based on the activity data collected by different sensors, such as a significant increase in the respiration rate of the user. The respiration rate may be collected by a respiration sensor.

[0054]

[0051] Further, the method 300 further includes determining a sleep inertia of the user based on the sleep end time and the active time, at step 308. The sleep inertia is determined by calculating a gap between the sleep end time and the active time. Quantitatively, the sleep inertia is the difference between the active time and the sleep end time of the user.

[0055]

[0052] The method 300 further includes calculating the morning alertness score of the user based on the sleep inertia, at step 310. The morning alertness score is calculated by a ML model implemented in one or more of the smart ring, user device, and network cloud. The ML model analyzes the sleep inertia, the heart rate, and the activity post active time to calculate a morning alertness score.

[0056]

[0053] Further, the method 300 further includes rendering the morning alertness score along with the activity data of the user via a user device, at step 312. The user device may include, but not limited to, a smart phone, a tablet, a laptop, a computer, etc. The ML model may calculate and render the insights on the morning alertness score providing users with actionable insights into their sleep quality and readiness for the day. Further, historical data of the user may be communicated to a network cloud or a decentralized database for performing trend analysis based on daily sleep inertia and morning alertness.

[0057]

[0054] One of the many technical advantages of the present invention include the capability to calculate the sleep end time of the user automatically based on the sleep pattern of the user. The invention provides the capability to calculate sleep inertia which corresponds to the time taken by the user to perform a significant activity post after waking up from sleep / sleep end time. The invention also provides the capability to determine the morning alertness of the user which may offer insights into sleep effectiveness and psychological markers such as motivation of the user.

Claims

WE CLAIM:

1. A wearable device, comprising: a PPG sensor (110) for collecting heart rate data of a user; a motion sensor (106) for collecting activity data of the user; a sleep tracking sensor (108) for determining a sleep end time of the user by analysing transitional states between different sleep stages of the user; and a Central Processing Unit (CPU) (102) communicatively coupled with a memory (104) for: calculating an active time of the user based on the predefined temporal windows immediately after detection of sleep end time and the comparing activity data, wherein the active time corresponds to a significant increase in activity after the sleep end time; determining an alertness score of the user based on the active time and the activity data, by analyzing quantity of activity performed by the user before the active time and after the sleep end time; and determining sleep inertia of the user by calculating a gap between the sleep end time and the active time.

2. The wearable device as claimed in claim 1, wherein the sleep tracking sensor (108) implements a sleep tracking algorithm for determining a wake-up time of the user.

3. The wearable device as claimed in claim 2, wherein for determining the wake-up time of the user, the sleep tracking algorithm identifies more than a predefined number of steps being taken within a predefined time period by the user.

4. The wearable device as claimed in claim 2, wherein for determining the wake-up time of the user, the sleep tracking algorithm monitors the wearable device being placed for charging.

5. The wearable device as claimed in claim 1, wherein the motion sensor (106) and the PPG sensor (110) keep collecting the activity of the user post sleep end time.

6. The wearable device as claimed in claim 1, wherein while determining the alertness score, the wearable device considers physiological parameters including heart rate, respiration rate, and oxygen volume.

7. The wearable device as claimed in claim 1, wherein the CPU (102) removes physiological outlier values for enhancing accuracy of the wearable device.

8. The wearable device as claimed in claim 1, wherein the morning alertness and connected sleep inertia is quantified via a mathematical function, wherein the mathematical function is one of slope, numerical difference, and vector difference.

9. The wearable device as claimed in claim 1, wherein the CPU (102) further determines a morning alertness score of the user based on the sleep inertia, the heart rate, and the activity post active time.

10. The wearable device as claimed in claim 1, wherein the wearable device is one of a smart ring (100), smart watch, and smart band.

11. The wearable device as claimed in claim 1, wherein the wearable device communicates historical data of the user to a network cloud or a decentralized database for performing trend analysis based on daily sleep inertia and morning alertness.

12. The wearable device as claimed in claim 1, wherein the wearable device is connected with a user device for rendering the morning alertness score along with the activity data of the user.

13. A method of determining morning alertness of a user, the method comprising: collecting, using a PPG sensor (110), heart rate data of a user; collecting, using a motion sensor (106), activity data of the user; determining, using a sleep tracking sensor (108), a sleep end time of the user by analysing transitional states between different sleep stages of the user; calculating, by a Central Processing Unit (CPU) (102), an active time of the user based on the sleep end time and the activity data, wherein the active time corresponds to a significant increase in activity after the sleep end time; anddetermining, by the CPU (102), an alertness score of the user based on the active time and the activity data, by analyzing quantity of activity performed by the user before the active time and after the sleep end time.

14. The method as claimed in claim 13, wherein the sleep tracking sensor (108) implements a sleep tracking algorithm for determining a wake-up time of the user.

15. The method as claimed in claim 14, wherein for determining the wake-up time of the user, the sleep tracking algorithm identifies more than a predefined number of steps being taken within a predefined time period by the user.

16. The method as claimed in claim 14, wherein for determining the wake-up time of the user, the sleep tracking algorithm monitors the wearable device being placed for charging.

17. The method as claimed in claim 13, wherein the motion sensor (106) and the PPG sensor (110) keep collecting the activity of the user post sleep end time.

18. The method as claimed in claim 13, wherein while determining the alertness score, the wearable device considers physiological parameters including heart rate, respiration rate, and oxygen volume.

19. The method as claimed in claim 13, wherein the gap indicative of the sleep inertia is quantified via a mathematical function, wherein the mathematical function is one of slope, numerical difference, and vector difference.

20. The method as claimed in claim 13, further comprising: determining sleep inertia of the user by calculating a gap between the sleep end time and the active time; determining a morning alertness score of the user based on the sleep inertia, the heart rate, and the activity post active time; and causing a graphical user interface to display an indication of the morning alertness.

Images