Reproductive Stage Coaching

Abstract

Physiological indicators such as respiration rate, resting heart rate, heart rate variability, and body temperature are measured over time for the user and correlated with reproductive stages. By identifying chronological stages such as hormonal cycles, automated recommendations regarding sleep, diet, exercise, etc. can be provided in a progressively tailored manner.

Description

[Technical Field]

[0001] [Related Applications] This application claims priority to U.S. Patent Application No. 63 / 404,247, filed September 7, 2022, the entire contents of which are incorporated herein by reference.

[0002] The present disclosure relates generally to providing recommendations and coaching based on reproductive stages such as menstrual cycle, pregnancy, and menopause. [Background technology]

[0003] [background] Reproductive phases can impact health, fitness, recovery, sleep, etc. There remains a need for technology that identifies reproductive phases in a way that facilitates providing tailored coaching and recommendations related to exercise, recovery, sleep, diet, and other areas of health and fitness. Summary of the Invention [Means for solving the problem]

[0004] [overview] Physiological indicators such as respiration rate, resting heart rate, heart rate variability, and body temperature can be measured over time for a user and may correlate with reproductive stages. Identifying temporal stages such as hormonal cycles allows for automated recommendations for sleep, diet, exercise, etc. to be tailored to the stage.

[0005] In one aspect, a computer program product disclosed herein may include computer-executable code embodied in a non-transitory computer-readable medium, which, when executed on one or more computing devices, may perform the following steps: providing a model characterizing changes over time in each of heart rate variability, resting heart rate, body temperature, and respiration rate during a model hormonal cycle; acquiring physiological data of a user from a wearable monitor, the physiological data including at least heart rate data and body temperature data, and the physiological data acquired during the user's hormonal cycle; calculating a plurality of metrics of the user at least daily during the hormonal cycle, the plurality of metrics including at least the heart rate variability, the resting heart rate, the body temperature, and the respiration rate; calculating an estimated cycle time for the user for the model hormonal cycle based on each of the plurality of metrics separately; calculating a cycle time within the hormonal cycle for the user based on the set of estimated cycle times; and providing coaching information to the user based on the cycle time.

[0006] Embodiments may include one or more of the following features: The hormonal cycle may include the user's menstrual cycle. The hormonal cycle may include the user's pregnancy. The set may include a weighted average of the estimated cycle times for each of the plurality of metrics. The set may include a combination of the estimated cycle times for each of the plurality of metrics based on a probability of accurately estimating the cycle time.

[0007] In one aspect, the method disclosed herein may include providing a model characterizing changes over time during a model hormone cycle for each of two or more physiological indicators; acquiring heart rate data from a wearable monitor worn by a user; calculating the two or more physiological indicators for the user based on the heart rate data, at least daily; calculating a cycle time within the user's hormone cycle based on a set of estimated cycle times, each estimated cycle time being derived by applying one of the physiological indicators to the model; and providing coaching information to the user based on the cycle times.

[0008] Embodiments may include one or more of the following features. The hormonal cycle may include the user's menstrual cycle. The hormonal cycle may include the user's pregnancy. The set may include a weighted average of estimated cycle times for each of the physiological indicators. The set may include a combination of the estimated cycle times based on a probability of accurately estimating the cycle time. The set may include a Bayesian model average of the estimated cycle times. The set may include an average of at least two of the estimated cycle times. The wearable monitor may include a photoplethysmography monitor. The two or more physiological indicators may include at least one of heart rate variability, resting heart rate, and respiratory rate. The two or more physiological indicators may include body temperature, and the wearable monitor may include a temperature sensor, and the method may include obtaining temperature data from the temperature sensor and calculating the user's body temperature at least daily. The model hormonal cycle may be derived from a population of users. The model hormonal cycle may be derived from the user's history.

[0009] In one embodiment, the system disclosed herein may include a wearable monitor configured to acquire heart rate data from a user, a model stored in memory, the model characterizing changes over time during a model hormone cycle for each of two or more physiological indicators, and a processor. The processor may be configured to generate recommendations for the user by performing the following steps: receiving the heart rate data from the wearable monitor; periodically calculating the two or more physiological indicators of the user based on the heart rate data; calculating a cycle time within the user's hormone cycle based on a set of estimated cycle times, each estimated cycle time being derived by applying one of the physiological indicators to the model; and providing coaching information to the user based on the cycle time. The processor may run on the user's personal computing device. The processor may run on a remote server connected to the wearable monitor via a data network.

[0010] In one aspect, a computer program product may include computer-executable code embodied in a non-transitory computer-readable medium, which, when executed on one or more computing devices, may perform the following steps: providing a model characterizing changes over time during a model hormone cycle for each of two or more physiological indicators, acquiring heart rate data from a wearable monitor worn by a user, calculating the two or more physiological indicators for the user at least daily based on the heart rate data, monitoring the user's hormone cycle by applying the two or more physiological indicators to the model hormone cycle, identifying one or more temporal irregularities of the hormone cycle relative to the model hormone cycle, and providing a recommendation to the user in response to an estimated likelihood of menopause onset exceeding a predetermined threshold based on the one or more temporal irregularities.

[0011] Embodiments may include one or more of the following features: The model may be derived from a population of users. The model may be based on the user's history. The two or more physiological indicators may include at least one of heart rate variability, resting heart rate, and respiration rate. The wearable monitor may include a temperature sensor, and the two or more physiological indicators may include body temperature, and the computer program product may include code for obtaining temperature data from the temperature sensor and calculating the body temperature of the user at least daily. Identifying the one or more temporal irregularities may include detecting deviations from the model in at least one of the physiological indicators. Identifying the one or more temporal irregularities may include detecting deviations from the model in a set of the two or more physiological indicators. Identifying the one or more temporal irregularities may include detecting changes in the expected length of the hormone cycle. The recommendations may include at least one of dietary recommendations, sleep recommendations, and activity recommendations. The wearable monitor may include a photoplethysmography monitor.

[0012] In one aspect, a computer program product disclosed herein may include computer-executable code embodied in a non-transitory computer-readable medium, which, when executed on one or more computing devices, may perform the following steps: providing a model characterizing changes over time during a model hormone cycle for each of two or more physiological indicators, the values ​​of which are affected by one or more hormones associated with the model hormone cycle; acquiring heart rate data from a wearable monitor worn by a user; calculating the two or more physiological indicators for the user at least daily based on the heart rate data; monitoring the user's hormone cycle by applying the two or more physiological indicators to the model hormone cycle; identifying, for each of the two or more physiological indicators, a series of peaks within the hormone cycle; identifying, for the series of peaks, a decrease in magnitude of each of the two or more physiological indicators over time; providing a prediction of menopausal onset for the user in response to the decrease in magnitude over time; and notifying the user of the predicted menopausal onset.

[0013] Embodiments may include one or more of the following features: The model may be derived from a population of users. The model may be based on the user's history. The two or more physiological indicators may include at least one of heart rate variability, resting heart rate, and respiration rate. The wearable monitor may include a temperature sensor, and the two or more physiological indicators may include body temperature, and the computer program product may include code that performs the steps of obtaining temperature data from the temperature sensor and calculating the body temperature of the user at least daily. The computer program product may further include code that, when executed on the one or more computing devices, provides recommendations to the user based on the prediction of menopausal onset, the recommendations including at least one of dietary recommendations, sleep recommendations, and activity recommendations. The wearable monitor may include a photoplethysmography monitor.

[0014] In one aspect, a system disclosed herein may include a wearable monitor configured to acquire heart rate data from a user and a processor. The processor may be configured to receive heart rate data from the wearable monitor, periodically calculate two or more physiological indicators of the user based on the heart rate data, the two or more physiological indicators having values ​​influenced by one or more hormones associated with the user's hormonal cycle, generate a prediction of menopausal onset for the user based on a predetermined pattern of the two or more physiological indicators over time, and provide coaching information to the user based on the prediction of menopausal onset. The hormonal cycle may be identified by applying the two or more physiological indicators to a hormonal cycle model, where the predetermined pattern may include one or more temporal irregularities in the hormonal cycle. The hormonal cycle may be identified by applying the two or more physiological indicators to a hormonal cycle model, where the predetermined pattern may include a decrease in magnitude of each of the two or more physiological indicators over time for a series of peaks in the hormonal cycle.

[0015] In one aspect, a computer program product for recommending adjustments to an activity plan based on reproductive stage as disclosed herein may include non-transitory computer-executable code embodied in a computer-readable medium that, when executed on one or more computing devices, may perform the following steps: acquire physiological data of a user from a wearable physiological monitoring device, identify a stage of the user's hormonal cycle based on the physiological data, determine a current recovery level of the user based on the user's past sleep activity, generate recommended goals for the user's activity plan based on the current recovery level, and automatically adjust the activity plan for the user by adjusting the recommended goals based on the stage of the hormonal cycle.

[0016] In one aspect, a system disclosed herein may include a wearable physiological monitoring device including one or more sensors, a first processor configured to substantially continuously acquire user heart rate data based on signals from the one or more sensors, and a communications interface for coupling with a remote resource; a server communicatively coupled to the wearable physiological monitoring device, the server including a second processor configured with computer-executable code to acquire physiological data of the user from the wearable physiological monitoring device, identify the user's reproductive stage based on the physiological data, determine the user's current recovery level based on the user's past sleep activity, generate recommended goals for the user's activity plan based on the current recovery level, and automatically adjust the user's activity plan by adjusting the recommended goals based on the reproductive stage; and a user interface configured to present the recommended goals to the user. The reproductive stage may include any one of a pregnancy trimester, a postpartum period, menopause, and perimenopause.

[0017] In one aspect, a method disclosed herein may include obtaining physiological data of a user from a wearable physiological monitoring device; identifying a reproductive stage of the user based on the physiological data; determining a current recovery level of the user based on the user's past sleep activity; generating recommended goals for an activity plan by the user based on the current recovery level; and automatically adjusting the activity plan of the user by adjusting the recommended goals based on the reproductive stage.

[0018] Embodiments may include one or more of the following features. The reproductive stage may include a pregnancy trimester. Identifying the reproductive stage may include identifying a gestational age of a fetus. The reproductive stage may include one of menopause and perimenopause. The physiological data may include heart rate data. Identifying the reproductive stage may include identifying the reproductive stage based on a pattern of change in the user's heart rate variability. Identifying the reproductive stage may include identifying a respiration rate of the user and identifying the reproductive stage based on a pattern of change in the user's respiration rate. Identifying the reproductive stage may include identifying the respiration rate based on the user's heart rate variability. Identifying the reproductive stage may include training a machine learning model to detect the reproductive stage based on one or more of the user's respiration rate and resting heart rate. The past sleep activity may be based on one or more of the user's past strain, heart rate variability, resting heart rate, and respiration rate. The user's past sleep activity may include sleep durations of past sleep events. The recommended goal may include a goal related to one or more of activity amount and activity intensity. The recommended goal may include a sleep goal. Adjusting the recommended goal may include adjusting a duration of the sleep goal. The method may include presenting the recommended goal to the user on a user interface. The method may include presenting the reproductive stage to the user on a user interface. The physiological data may be collected substantially continuously by a wearable physiological monitoring device. [Brief explanation of the drawings]

[0019] These and other objects, features, and advantages of the devices, systems, and methods described herein will become apparent from the following description of specific embodiments thereof, as illustrated in the accompanying drawings. The drawings are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the devices, systems, and methods described herein. Like reference numbers throughout the drawings generally identify corresponding elements.

[0020] [Figure 1] FIG. 1 illustrates a physiological monitoring device.

[0021] [Figure 2] FIG. 1 illustrates a physiological monitoring system.

[0022] [Figure 3] FIG. 1 illustrates a smart clothing system.

[0023] [Figure 4] FIG. 1 is a block diagram of a computing device.

[0024] [Figure 5] FIG. 1 illustrates a dynamic stress monitoring system.

[0025] [Figure 6] 1 is a flow diagram illustrating a signal processing algorithm for generating a sequence of heart rates for each detected heart beat, which may be embodied in computer-executable instructions stored on one or more non-transitory computer-readable media.

[0026] [Figure 7] FIG. 1 is a flow diagram illustrating a method for determining an intensity score.

[0027] [Figure 8] FIG. 1 is a flow diagram illustrating how a user can use the intensity and recovery scores.

[0028] [Figure 9] 1 illustrates a display of an intensity score index shown as a circular graphic component, illustrating the current score as 19.0.

[0029] [Figure 10] FIG. 10 illustrates a display of the recovery score index shown in a circular graphic component with a first threshold of 66% and a second threshold of 33% indicated.

[0030] [Figure 11A] FIG. 10 illustrates a recovery score graphic component that includes a recovery score and qualitative information corresponding to the recovery score.

[0031] [Figure 11B] FIG. 10 illustrates a recovery score graphic component that includes a recovery score and qualitative information corresponding to the recovery score.

[0032] [Figure 11C] FIG. 10 illustrates a recovery score graphic component that includes a recovery score and qualitative information corresponding to the recovery score.

[0033] [Figure 12A] FIG. 1 illustrates a portion of a user interface for displaying user-specific physiological data displayed on a visual display device.

[0034] [Figure 12B] FIG. 1 illustrates a portion of a user interface for displaying user-specific physiological data displayed on a visual display device.

[0035] [Figure 13A] FIG. 1 illustrates a portion of a user interface for displaying user-specific physiological data displayed on a visual display device.

[0036] [Figure 13B] FIG. 1 illustrates a portion of a user interface for displaying user-specific physiological data displayed on a visual display device.

[0037] [Figure 14A] FIG. 1 illustrates a portion of a user interface for displaying user-specific physiological data displayed on a visual display device.

[0038] [Figure 14B] FIG. 1 illustrates a portion of a user interface for displaying user-specific physiological data displayed on a visual display device.

[0039] [Figure 15] FIG. 10 is a flow diagram illustrating a method for selecting a mode for acquiring heart rate data.

[0040] [Figure 16] FIG. 1 is a flow diagram of a method for assessing recovery and recommending exercise.

[0041] [Figure 17] FIG. 1 is a flow diagram illustrating a method for detecting heart rate variability during sleep.

[0042] [Figure 18] FIG. 1 is a flow diagram illustrating a method for detecting sleep intention.

[0043] [Figure 19] FIG. 1 is a flow diagram illustrating a method for recommending adjustments to an activity plan based on reproductive stage.

[0044] [Figure 20A] FIG. 1 illustrates correlations useful for automatically detecting menstrual cycles.

[0045] [Figure 20B] FIG. 1 illustrates correlations useful for automatically detecting menstrual cycles.

[0046] [Figure 21] FIG. 1 is a flow diagram illustrating a method for recommending strain-related adjustments.

[0047] [Figure 22] FIG. 1 is a flow diagram illustrating a method for recommending fitness and nutrition related adjustments.

[0048] [Figure 23] FIG. 1 is a flow diagram illustrating a method for recommending sleep-related adjustments based on a stage within a menstrual cycle.

[0049] [Figure 24] FIG. 1 is a flow diagram illustrating a method for recommending sleep-related adjustments based on pregnancy trimester.

[0050] [Figure 25] FIG. 1 is a flow diagram illustrating a method for recommending sleep-related adjustments based on menopausal or perimenopausal stage.

[0051] [Figure 26] FIG. 1 is a flow diagram of a method for providing coaching recommendations based on hormone cycles.

[0052] [Figure 27] FIG. 1 illustrates a model of the menstrual cycle.

[0053] [Figure 28] FIG. 1 illustrates a model of the pregnancy cycle.

[0054] [Figure 29] FIG. 1 is a flow diagram of a method for detecting the onset of menopause. DETAILED DESCRIPTION OF THE INVENTION

[0055] [explanation] Various embodiments will now be described in more detail below with reference to the accompanying drawings, in which preferred embodiments are shown. However, the above should not be construed as being limited to the embodiments described herein, as they may be embodied in many different forms. Rather, the illustrated embodiments are provided so that this disclosure will convey its scope to those skilled in the art.

[0056] All documents mentioned herein are incorporated herein by reference in their entirety. Reference to a singular item should be understood to include the plural item, and vice versa, unless expressly stated otherwise or apparent from the context. Grammatical conjunctions are intended to express all disjunctive and conjunctive combinations of joined clauses, sentences, words, etc., unless stated otherwise or apparent from the context. Thus, the word "or" should generally be understood to mean "and / or," etc.

[0057] The description of ranges of values ​​herein is not intended to be limiting and, unless otherwise specified, refers individually to every value within that range, and each value within such range is incorporated herein as if it were individually set forth herein. Words such as "about," "approximately," and the like associated with numerical values ​​are intended to indicate a degree of deviation that one skilled in the art would recognize as sufficient for the intended or stated purpose. Similarly, approximation words such as "approximately" or "substantially" used in connection with physical properties should be interpreted as indicating a range of deviation that one skilled in the art would recognize as sufficient for the corresponding use, function, purpose, etc. Value ranges and / or numerical values ​​are provided herein as examples only and do not limit the scope of the described embodiments. When ranges of values ​​are presented, unless expressly stated otherwise, it is intended that the ranges include each value within the range as if it were individually set forth. The use of any examples or exemplary language (such as "for example," "e.g.," etc.) provided herein is intended only to better illustrate the embodiments and does not limit the scope of the disclosed embodiments. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the embodiments.

[0058] In the following description, it will be understood that terms such as "first," "second," "upper," "lower," "above," "below," "upper," "under," etc. are terms of convenience and should not be construed as limiting terms unless specifically stated otherwise.

[0059] Exemplary embodiments provide physiological measurement systems, devices, and methods for continuous health and fitness monitoring, offering improvements to overcome the shortcomings of conventional heart rate monitors. One aspect of the present disclosure is directed to providing a lightweight, wearable system with a strap that collects various physiological data or signals from a wearer. The strap may be used to attach the system to a user's body part or limb, such as the wrist or ankle. The exemplary system is wearable and enables real-time, continuous monitoring of heart rate without the need for a chest strap or other bulky equipment, which can cause discomfort and discourage continuous wear and use. The system can determine a user's heart rate without the use of an electrocardiogram or the need for a chest strap. Thus, the exemplary system can be used for continuous fitness monitoring as well as for assessing general health. In addition to heart rate, the exemplary system may also monitor one or more physiological parameters, including, but not limited to, body temperature, heart rate variability, movement, sleep, stress, fitness level, recovery level, the effect of a workout routine on health and fitness, calorie expenditure, etc.

[0060] Health or fitness monitors that include bulky components can be a hindrance to continuous wear. Existing fitness monitors often incorporate watch functionality, making the health or fitness monitor very bulky and inconvenient for continuous wear. Accordingly, one aspect is directed to providing a wearable health or fitness system that does not include bulky components, thereby making the bracelet slimmer, less obtrusive, and more suitable for continuous wear. The ability to wear the bracelet continuously further enables continuous collection of physiological data and continuous, more reliable monitoring of health or fitness. For example, embodiments of the bracelet disclosed herein allow a user to monitor data constantly, not just during a fitness session. In some embodiments, the wearable system may or may not include a display screen for displaying heart rate and other information. In other embodiments, the wearable system may include one or more light-emitting diodes (LEDs) for providing feedback to the user and selectively displaying heart rate. In some embodiments, the wearable system may include a removable or openable modular head that can provide additional functionality or display additional information. Such a modular head may be removably attached to the wearable system when additional information display is desired, and may be removed to improve the comfort and appearance of the wearable system. In other embodiments, the head may be integrally formed with the wearable system.

[0061] Exemplary embodiments further include computer-executable instructions that, when executed, enable automated interpretation of one or more physiological parameters to assess a user's experienced cardiovascular intensity (embodied in an intensity score or index) and the user's recovery after physical exercise or daily stress through sleep and other forms of rest (embodied in a recovery score). These indexes or scores may be stored and displayed in a meaningful format to help users manage their health and exercise regimen. Exemplary computer-executable instructions may be provided in a cloud embodiment.

[0062] Exemplary embodiments also provide a vibrant, interactive online community in the form of a website for viewing and sharing physiological data among users. Website users include individuals receiving health and fitness monitoring, such as individuals wearing the wearable system disclosed herein, athletes, sports team members, personal trainers, coaches, etc. In some embodiments, users can select their trainer from a list to comment on their performance. The exemplary system has the capability to wirelessly stream all physiological information to an online website, either directly or via a mobile communication device application, using data transfer to a mobile phone / computer. This information, and the data described herein, may be encrypted (e.g., the data may include encrypted biometric data). Accordingly, the encrypted data may be streamed to a secure server for processing. In this way, only authorized users can view the data and associated scores. Additionally or alternatively, the website allows users to monitor their fitness results, share information with teammates and coaches, compete against other users, and earn status. Both the wearable system and the website allow users to get feedback about their day, exercise, and / or sleep, allowing for recovery and performance assessment.

[0063] In one exemplary technique for data transmission, data collected by the wearable system is transmitted directly to cloud-based data storage, from which the data can be downloaded for display and analysis on a website. In another exemplary technique for data transmission, data collected by the wearable system is transmitted via a mobile communication device application to cloud-based data storage, from which the data can be downloaded for display and analysis on a website.

[0064] In some embodiments, the website may be a social networking site. In some embodiments, the website may be displayed using a mobile website or a mobile application. In some embodiments, the website may be configured to communicate data to other websites or applications. In some embodiments, the website may be configured to provide an interactive user interface. The website may be configured to display results based on an analysis of physiological data received from one or more devices. The website may be configured to provide a competitive way for users to compare with each other, ultimately providing a more interactive experience for users. For example, in some embodiments, rather than simply comparing a user's physiological data and performance with that user's past performance, a user may be able to compete against other users, and a user's performance may be compared with that of other users.

[0065] To facilitate understanding of the illustrative embodiments, certain terms are defined below.

[0066] As used herein, the term "user" refers to any type of animal, human or non-human, having physiological information that can be monitored using the exemplary wearable physiological monitoring system.

[0067] As used herein, the term "body" refers to the body of the user.

[0068] The term "continuous" as used herein with respect to heart rate data refers to acquiring heart rate data frequently enough to detect individual heartbeats, and also refers to collecting heart rate data over extended periods of time, such as an hour, a day, or longer (including acquisition throughout the day and night). More generally, "continuous" or "continuously," with respect to physiological signals that may be monitored by a wearable device, is understood to mean continuous at a rate and duration appropriate for intended time-based processing, and physically, at a periodic rate sufficient to resolve desired physiological characteristics, such as heart rate, heart rate variability, heart rate peak detection, pulse shape, etc. (e.g., multiple times per heartbeat, breath, etc.). At the same time, continuous monitoring is not intended to exclude interruptions in normal data acquisition, such as sudden movement, changes in external lighting, loss of power, physical manipulation and / or adjustment by the wearer, temporary movement of the monitoring hardware due to external forces, etc. It should also be noted that heart rate data or monitored heart rate in this context may more generally refer to raw sensor data, such as light intensity signals, or to processed data, such as heart rate data, signal peak data, heart rate variability data, or other physiological or digital signals suitable for recovering heart rate information as contemplated herein. Furthermore, such heart rate data is typically obtained over a historical period and may then be correlated with various other data or metrics related to, for example, sleep state, perceived athletic activity, resting heart rate, maximum heart rate, etc.

[0069] As used herein, the term "pointing device" refers to any suitable input interface, particularly a human interface device, that allows a user to input spatial data into a computing system or device. In an exemplary embodiment, a pointing device allows a user to provide input to a computer using physical gestures such as, for example, pointing, clicking, dragging, and dropping. Exemplary pointing devices may include, but are not limited to, a mouse, a touchpad, a touchscreen, and the like.

[0070] As used herein, the term "multi-chip module" refers to an electronic package in which multiple integrated circuits (ICs) are packaged on a unified substrate so that they can be used as a single component - i.e., high-performance ICs packaged in a very small volume.

[0071] As used herein, the term "computer-readable medium" refers to a non-transitory storage medium, such as storage hardware, storage devices, or computer memory, accessible by a controller, microcontroller, microprocessor, computing system, or the like, or other modules or components or modules of a computing system, capable of encoding computer-executable instructions, software programs, and / or other data. A "computer-readable medium" may be accessed by a computing system or a module of a computing system to retrieve and / or execute computer-executable instructions or software programs encoded therein. Non-transitory computer-readable media may include, but are not limited to, one or more types of hardware memory, non-transitory tangible media (e.g., one or more magnetic storage disks, one or more optical disks, one or more USB flash drives), virtual or physical computer system memory, physical memory hardware such as random access memory (DRAM, SRAM, EDORAM, etc.), and the like. Although not shown, any device or component described herein may include a computer-readable medium or other memory for storing program instructions, data, and the like.

[0072] As used herein, the term "distal" refers to the part, end, or component of the physiological measurement system that is furthest from the user's body when worn by the user.

[0073] As used herein, the term "proximal" refers to the part, end, or component of the physiological measurement system that is closest to the user's body when worn by the user.

[0074] The term "equal" as used herein is used in a broad sense to refer to exact equivalence or approximate equivalence within some tolerance.

[0075] I. Exemplary Wearable Physiological Measurement System

[0076] An exemplary embodiment provides a wearable physiological measurement system configured to continuously measure heart rate. The exemplary system is configured to be continuously worn on a body part, such as the wrist or ankle, and does not rely on an electrocardiogram or chest strap to detect heart rate. The exemplary system includes one or more light emitters that emit light of one or more desired frequencies toward a user's skin and one or more photodetectors that receive light reflected from the user's skin. The photodetectors may include photoresistors, phototransistors, photodiodes, etc. When light from the light emitters (e.g., green light) penetrates the user's skin, the photoresistor reading fluctuates depending on blood's natural absorption or transmittance of light. These waves have the same frequency as the user's pulse, because increased absorption or transmittance only occurs when blood flow increases after a heartbeat. The system includes a processing module, implemented in software, hardware, or a combination thereof, for processing optical data received by the photodetectors and continuously determining the heart rate based on the optical data. The optical data can be combined with data from one or more motion sensors, such as an accelerometer or gyroscope, to minimize or eliminate noise in the heart rate signal caused by movement or other artifacts (alternatively, the optical data may be combined with other optical data at different wavelengths).

[0077] FIG. 1 illustrates a physiological monitoring system. The system 100 may include a wearable monitor 104 configured for physiological monitoring. The system 100 may further include a removable and replaceable battery 106 for recharging the wearable monitor 104. The wearable monitor 104 may include a strap 102 or other retention system(s) for securing the wearable monitor 104 to a specific location on the wearer's body to acquire physiological data as described herein. For example, the strap 102 may include a thin elastic band formed of any suitable elastic material, such as rubber, or a woven polymer fiber, such as polyester, polypropylene, nylon, or spandex. The strap 102 may be adjustable to accommodate various wrist sizes and may include a latch, clasp, or the like for securing the wearable monitor 104 in a desired location for monitoring physiological signals. While a wrist-worn device is illustrated, it will be understood that the wearable monitor 104 may be configured for placement at any suitable location on the user's body based on the sensing means and nature of the signals to be acquired. For example, the wearable monitor 104 may be configured for use on the wrist, ankle, biceps, chest, or other suitable location, and the strap 102 may be or include a waistband or other elastic band within a garment or accessory. The wearable monitor 104 may also or alternatively be structurally configured for placement on or within a garment, e.g., permanently or in a removable and replaceable manner. To this end, the wearable monitor 104 may be shaped and sized to be placed within a pocket, slot, and / or other housing coupled to or embedded within the garment. In such a configuration, the garment pocket or other retention means may include a sensing window or the like so that the wearable monitor 104 can operate even when used within the garment.Non-limiting examples of suitable wearable monitor 104 embodiments are described in US Pat. No. 11,185,292, which is incorporated herein by reference in its entirety.

[0078] System 100 may include any hardware components, subsystems, etc. for supporting various functions of wearable monitor 104, such as data collection, processing, display, and communication with external resources. For example, system 100 may include hardware for heart rate monitoring using photoplethysmography (PPG), electrocardiography, or other techniques. System 100 may be configured such that when wearable monitor 104 is worn on the wrist (or elsewhere on the body) during use, system 100 begins acquiring physiological data from the wearer. In some embodiments, pulse and / or rate may be acquired optically using a light source (e.g., a light-emitting diode (LED)) and a photodetector within wearable monitor 104. The LED may be positioned to direct light toward the user's skin, and a photodetector, such as a photodiode, may be used to acquire light intensity measurements indicative of light from the LED reflected and / or transmitted by the wearer's skin.

[0079] System 100 may be configured to record other physiological and / or biomechanical parameters, including, but not limited to, skin temperature (using a thermometer), galvanic skin response (GSR) (using a galvanic skin response sensor), movement (using one or more multi-axis accelerometers and / or gyroscopes), blood pressure, etc., as well as environmental or situational parameters, such as ambient light, ambient temperature, humidity, time of day, etc. For example, wearable monitor 104 may include sensors such as an accelerometer and / or gyroscope for motion detection, sensors for sensing ambient temperature, sensors for measuring electrodermal activity (EDA), sensors for measuring galvanic skin response (GSR) detection, etc. System 100 may additionally or alternatively include other systems or subsystems that support additional functionality of wearable monitor 104. For example, system 100 may include a communication system supporting near field communication (NFC), proximity sensing, Bluetooth® communication, Wi-Fi communication, cellular communication, satellite communication, etc. The wearable monitor 104 may additionally or alternatively include components such as a global positioning system (GPS), a display and / or user interface, a clock and / or timer, etc.

[0080] The wearable monitor 104 may include one or more battery power sources, such as a first battery within the wearable monitor 104 and a second battery 106 that is removable and replaceable from the wearable monitor 104 to recharge the battery within the wearable monitor 104. The system 100 may additionally or alternatively include multiple wearable monitors 104 (and / or other physiological monitors) that can share battery power or power each other. The system 100 may perform numerous functions related to continuous monitoring, such as automatically detecting whether the user is asleep, awake, exercising, etc. These detections may be performed locally on the wearable monitor 104 or by a remote service communicatively coupled to and receiving data from the wearable monitor 104. In general, the system 100 may support continuous and independent monitoring of physiological signals, such as heart rate, and the underlying acquired data may be stored on the wearable monitor 104 for an extended period of time before being uploaded to a remote processing resource for more complex computational analysis.

[0081] In one embodiment, the wearable monitor may be a wrist-worn photoplethysmography device.

[0082] 2 illustrates a physiological monitoring system. More specifically, FIG. 2 illustrates a physiological monitoring system 200 that may be used in any of the methods or apparatus described herein. Generally, the system 200 may include a physiological monitor 206, a user device 220, a remote server 230 with remote data processing resources (such as any of the processors or processing resources described herein), and one or more other resources 250, all of which may be interconnected via a data network 202.

[0083] Data network 202 may be any of the various data networks described herein. For example, data network 202 may be any network or internetwork suitable for communicating data or information between participants in system 200. This may include public networks such as the Internet, private networks, telecommunications networks such as the Public Switched Telephone Network (PSTN), or cellular networks using third-generation (e.g., 3G or IMT-200), fourth-generation (e.g., LTE (E-UTRA) or WiMAX-Advanced (IEEE 802.16m)), fifth-generation (e.g., 5G), and / or other technologies, as well as various enterprise or local area networks and other switches, routers, hubs, gateways, etc., that may be used to transmit data between participants in system 200. This may further include local or short-range communications infrastructure suitable, for example, for coupling physiological monitor 206 to user device 220 or for supporting communication with local resources. By way of non-limiting example, short-range communication may include Wi-Fi communication, Bluetooth communication, infrared communication, NFC (near field communication), communication with an RFID tag or reader, and the like.

[0084] The physiological monitor 206 may generally be any physiological monitoring device or system, such as any of the wearable monitors or other monitoring devices or systems described herein. In one embodiment, the physiological monitor 206 may be a wearable physiological monitor shaped and sized to be worn on the wrist or other body part. The physiological monitor 206 may include a wearable housing 211, a network interface 212, one or more sensors 214, one or more light sources 215, a processor 216, a haptic device 217 or other user input / output hardware, memory 218, and a strap 210 for holding the physiological monitor 206 in a desired position on the user. In one embodiment, the physiological monitor 206 may be configured to acquire heart rate data and / or other physiological data from the wearer intermittently or substantially continuously. In another embodiment, the physiological monitor 206 may be configured to support long-term, continuous acquisition of physiological data, for example, over a period of several days, a week, or more.

[0085] The network interface 212 of the physiologic monitor 206 may be configured to communicatively couple the physiologic monitor 206 to one or more other components of the system 200. This communication may be direct, such as via a cellular data connection, or indirectly via a short-range wireless communication channel. The short-range wireless communication channel locally couples the physiologic monitor 206 to a wireless access point, router, computer, laptop, tablet, mobile phone, or other device. These devices may process data locally and / or relay data from the physiologic monitor 206 to a remote server 230 or other resource(s) 250 depending on the need or usefulness for obtaining and processing data from the physiologic monitor 206.

[0086] The one or more sensors 214 may include any of the sensors described herein or other sensors or subsystems suitable for physiological monitoring or support functions. By way of example and not limitation, the one or more sensors 214 may include one or more of a light source, a light sensor, an accelerometer, a gyroscope, a temperature sensor, a galvanic skin response sensor, a capacitance sensor, a resistive sensor, an environmental sensor (e.g., for measuring ambient temperature, humidity, lighting, etc.), a geolocation sensor, a global positioning system (GPS), a proximity sensor, an RFID tag reader, an RFID tag, a time sensor, an electrodermal activity sensor, etc. The one or more sensors 214 may be located within the wearable housing 211 or may be located elsewhere and configured to provide physiological monitoring or other functions described herein. In one embodiment, the one or more sensors 214 may include a photodetector configured to provide light intensity data to the processor 216 (or to the remote server 230) for calculating heart rate and heart rate variability. The one or more sensors 214 may additionally or alternatively include an accelerometer, a gyroscope, etc. configured to provide motion data to the processor 216, for example, to detect activities such as sleep state, rest state, wake-up events, movement, and / or other user activity. In one embodiment, the one or more sensors 214 may include a sensor for measuring the user's electrodermal response. The one or more sensors 214 may additionally or alternatively include electrodes, etc. for capturing electrical signals, for example, to obtain an electrocardiogram and / or other electrically derived physiological measurements.

[0087] The processor 216 and memory 218 may be any of the processors and memories described herein. In one embodiment, the memory 218 may store physiological data obtained by monitoring the user using one or more sensors 214, and / or other sensor data, program data, or other data useful in the operation of the physiological monitor 206 or other components of the system 200. While only memory 218 is shown in the physiological monitor 206, it will be understood that any other device or component of the system 200 may additionally or alternatively include memory for storing program instructions, raw data, processed data, user input, etc. In one embodiment, the processor 216 of the physiological monitor 206 may be configured to obtain heart rate data from the user, for example, heart rate data including or based on raw data from the sensors 214. The processor 216 may also or alternatively be configured to identify or assist in identifying a user condition, for example, related to health, fitness, strain, restorative sleep, or any of the other conditions described herein.

[0088] One or more light sources 215 may be coupled to the wearable housing 211 and controlled by the processor 216. At least one of the light sources 215 is directed toward the user's skin adjacent the wearable housing 211. Light from the light sources 215, or more generally, light of one or more wavelengths of the light sources 215, may be detected by one or more of the sensors 214 and may be processed by the processor 216 as described herein.

[0089] System 200 may further include a remote data processing resource executing on remote server 230. The remote data processing resource may include any of the processors and associated hardware described herein and may be configured to receive data transmitted from memory 218 of physiological monitor 206 and process the data to detect or infer physiological signals of interest, such as heart rate, heart rate variability, respiratory rate, blood oxygen saturation, blood pressure, etc. Remote server 230 may additionally or alternatively assess user status, such as recovery status, sleep status, exercise activity, exercise type, sleep quality, daily activity strain, or any other health or fitness status that may be detected based on such data.

[0090] The system 200 may include one or more user devices 220. The user device 220 may cooperate with the physiological monitor 206 to, for example, provide a display for user data and analysis, or more generally provide user input / output, and / or provide a communications bridge from the network interface 212 of the physiological monitor 206 to the data network 202 and the remote server 230. For example, the physiological monitor 206 may locally communicate with the user device 220, such as a user's smartphone, via short-range communication, such as Bluetooth, to exchange data between the physiological monitor 206 and the user device 220. The user device 220 may then communicate with the remote server 230 via the data network 202 to transfer data from the physiological monitor 206 and receive analysis results from the remote server 230 for presentation to the user. In one embodiment, the user device(s) 220 may support physiological monitoring by processing, or pre-processing, data from the physiological monitor 206 to extract heart rate data and heart rate variability data from the raw data acquired by the physiological monitor 206. In another embodiment, the remote server 230 may have greater memory capacity and processing power than the physiological monitor 206 and / or the user device 220, and computationally intensive processing may be advantageously performed on the remote server 230.

[0091] The user device 220 may include any suitable computing device, including, but not limited to, a smartphone, a desktop computer, a laptop computer, a network computer, a tablet, a mobile device, a personal digital assistant (PDA), a mobile phone, a portable media or entertainment device, or any other computing device described herein. The user device 220 may include a user interface 222 to enable user access and analysis of data and / or to support user control of the operation of the physiological monitor 206. The user interface 222 may be managed by one or more applications executing locally on the user device 220. Alternatively, the user interface 222 may be provided and displayed remotely on the user device 220, for example, from a remote server 230 or one or more other resources 250.

[0092] Generally, the remote server 230 may include data storage, a network interface, and / or other processing circuitry. The remote server 230 may process data from the physiological monitor 206, perform physiological and / or health monitoring / analysis or other analyses described herein (e.g., sleep analysis, strain determination, recovery assessment, etc.), and may include a user interface for remotely accessing this data (e.g., from the user device 220). The remote server 230 may include a web server or other programmatic front end to facilitate web-based access by the user device 220 or physiological monitor 206 to the functions of the remote server 230 and other components of the system 200.

[0093] System 200 may include other resources 250, such as any resources that can be usefully used in the devices, systems, and methods described herein. For example, these other resources 250 may include other data networks, databases, processing resources, cloud data storage, data mining tools, computational tools, data monitoring tools, algorithms, etc. In another embodiment, other resources 250 may include one or more administrative or programmatic interfaces through which a human actor, such as a programmer, researcher, annotator, editor, analyst, or coach, can interact with any of the above. Other resources 250 may also or instead include any other software or hardware resources that can be usefully used in the network applications contemplated herein. For example, other resources 250 may include a payment processing server or platform used to authorize payments for access, content, or option / feature purchases. In another embodiment, other resources 250 may include a certificate server or other security resource for a third party to verify identity or encrypt or decrypt data. In another embodiment, other resources 250 may include user device 220, wearable strap 210, or a desktop computer co-located with (e.g., on the same local area network or directly connected thereto via a serial or USB cable) remote server 230. In this case, other resources 250 may provide ancillary functions to various components of system 200, such as firmware upgrades, a user interface, and storage and / or pre-processing of data from physiologic monitor 206 before transmission to remote server 230.

[0094] Other resources 250 may additionally or alternatively include one or more web servers that enable web-based access to and from other participants in system 200. While other resources 250 (e.g., web servers) are depicted as separate network entities, it will be readily understood that other resources 250 (e.g., web servers) may additionally or alternatively be logically and / or physically associated with one of the other devices described herein. Other resources 250 (e.g., web servers) may also include or provide a user interface 222 for web access to remote server 230 or database or other resource(s) to facilitate user interaction with information (e.g., from physiological monitor 206 or user device 220) over data network 202.

[0095] In another embodiment, other resources 250 may include fitness equipment or other fitness infrastructure. For example, a strength training machine may automatically record the number of repetitions and / or the weight added during repetitions. These may be wirelessly accessible by physiological monitors 206 or other user devices 220. More generally, a gym may be configured to track a user's movement between machines and receive activity reports from each machine to track various strength training activities during a workout. Other resources 250 may also or instead include other monitoring equipment or infrastructure. For example, system 200 may include one or more cameras for tracking the movement of free weights and / or the user's body position during repetitive movements, such as strength training activities. Similarly, a user may wear or have embedded in their clothing tracking fiducial markers, such as visually identifiable objects for image-based tracking or wireless beacons for other tracking. In another embodiment, the weights themselves may be equipped with devices, such as sensors for recording and communicating detected movements and / or beacons for self-identification of type, weight, etc., to facilitate automatic detection and tracking of athletic activity by other connected devices.

[0096] One limitation of wearable sensors is their placement on the body. Devices are typically worn on the wrist, occupying space that users may want to reserve for other devices or jewelry, or that may be aesthetically or functionally unadorned. This location also places limitations on what can be measured and may restrict the user's activities. For example, a user may be unable to wear boxing gloves while wearing a sensing device on their wrist. To address this issue, physiological monitors may also or instead be embedded in clothing. To accommodate physiological monitoring, clothing may be supplemented with communication interfaces, power sources, device location sensors, environmental sensors, geolocation hardware, payment processing systems, and other components that provide infrastructure and functionality extensions for the wearable physiological monitor. Such "smart clothing" may provide additional space on the user's body to support monitoring hardware and enable additional sensing technologies not possible with a single sensing device. Smart clothing also frees up body surface space for other devices.

[0097] As used herein, "smart garment" is generally understood to include garments incorporating infrastructure and devices to support, enhance, or complement various physiological monitoring modes. Such garments may include a wired local communication bus for in-garment hardware communication, a wireless communication system for in-garment hardware communication, a wireless communication system for communication outside the garment, etc. Garments may also include power sources, power management systems, processing hardware, data storage, etc., any of which may support the enhanced functionality of the smart garment.

[0098] 3 illustrates a smart garment system. Generally, the system 300 may include multiple components (e.g., garment 310, one or more modules 320, controller 330, processor 340, memory 342, etc.) that can communicate with each other via a data network 302. The garment 310 may be wearable by a user 301 and may be configured to communicate with a module 320 having a physiological sensor 322 structurally configured to sense a physiological parameter of the user 301. As described herein, the module 320 may be controllable by the controller 330 based at least in part on a location 316 where the module 320 is located on or within the garment 310. This location-based information may be derived through interaction and / or communication between the module 320 and the garment 310 using various techniques. Although two controllers 330 are shown, it is understood that the garment 310 may include a single inter-garment controller, or any number of garments 310 may include any number of individual controllers 330 (e.g., one per garment, or one for all garments worn by a person, etc.), and / or the controllers may be integrated into other modules 320.

[0099] For communication over the data network 302, the system 300 may include a network interface 304. The network interface 304 may be integrated into the garment 310, included in the controller 330, included in another module or component of the system 300, or a combination thereof. The network interface 304 may generally include any combination of hardware and software configured to wirelessly communicate data to remote resources. For example, the network interface 304 may connect to a laptop, smartphone, or the like using a local connection, and then connect to a wide area network to access web-based or other network-accessible resources. The network interface 304 may additionally or alternatively be configured to connect to a local access point, such as a router or wireless access point, to connect to the data network 302. In another aspect, the network interface 304 may be a cellular communication data connection for a direct wireless connection to a cellular network, or the like.

[0100] The data network 302 may be any of those described herein. By way of example, some embodiments of the system 300 may be configured to wirelessly stream information to a social network, a data center, a cloud service, or the like. In some embodiments, data streamed from the system 300 to the data network 302 may be accessible to the user 301 (or other users) via a website. Accordingly, the network interface 304 may be configured to wirelessly stream data collected by the system 300 to a remote processing facility 350, a database 360, and / or a server 370 for processing and access by the user. In some embodiments, the data may be transmitted automatically without user intervention. This may be done, for example, by storing the data locally and transmitting the data via available local area network resources, if a local access point, such as a wireless access point, or a relay device (e.g., a laptop, tablet, or smartphone) is available. In some embodiments, the system 300 may include a cellular system or other hardware for independently accessing network resources from the garment 310, without requiring a local network connection. As will be appreciated, the network interface 304 may include a computing device, such as a mobile phone. The network interface 304 may additionally or instead include, or be included on, another component of the system 300, or a combination thereof. Where this advantageously conserves battery power or communication resources, the system 300 may prioritize the use of local network resources, if available, and may reserve cellular communication for situations where the data storage capacity of the garment 310 is reached.Thus, for example, garment 310 may store data locally up to a predetermined local data storage threshold, below which it may transmit data over a local network if available, and garment 310 may transmit data to a central resource using a cellular data network only if local storage of data exceeds the predetermined threshold.

[0101] The garment 310 may include one or more designated areas 312 for placement of a module for detecting a physiological parameter of a user 301 wearing the garment 310. One or more of the designated areas 312 may be specially configured to receive a module 320 therein or thereon. For example, the designated area 312 may include a pocket structurally configured to accommodate the module 320. Additionally or alternatively, the designated area 312 may include a first fastener configured to cooperate with a second fastener disposed on the module 320. One or more of the first fastener and the second fastener may include at least one of a hook-and-loop fastener, a button, a clamp, a clip, a snap, a protrusion, and a cavity.

[0102] By placing a pocket or the like in one of these designated areas 312, the location of module 320 can be controlled. Using an RFID tag, sensor, or the like, designated area 312 can specifically detect when module 320 has been placed therein for monitoring purposes and can communicate the detected location to any appropriate control circuitry.

[0103] The garment 310 may additionally or alternatively incorporate other infrastructure 315 for cooperation with the module 320. For example, the garment infrastructure 315 may include infrastructure 315 associated with an ECG device, such as ECG pads (or conductive sensor pads and / or electrodes connecting to the module 320, the controller 330, and / or other components of the system 300) and leads. As yet another example, the garment infrastructure 315 may include wires embedded in the garment 310 to facilitate wired data or power transmission between the attached module 320 and other system components (including other modules 320). The infrastructure 315 may additionally or alternatively include various built-in functions, such as providing power to the modules, supporting data communication between the modules, and supporting operation of the system 300. Infrastructure 314 may additionally or alternatively include location tags or hardware, identification tags or hardware, a power source for powering modules 320 or other hardware, a communications infrastructure as described herein, an in-garment wired network, or auxiliary components (e.g., a processor, a Global Positioning System (GPS), a timing device for synchronizing signals from, e.g., multiple garments, a beacon for synchronizing signals between multiple modules 320, etc.) More generally, any hardware, software, or combination thereof suitable for enhancing the operation of garment 310 and physiological monitoring systems using garment 310 may be incorporated into garment 310, contemplated herein as infrastructure 315.

[0104] The modules 320 generally have a size and shape that allows them to be positioned on or within one or more designated areas 312 of the garment 310. For example, in some embodiments, one or more of the modules 320 may be permanently secured on or within the garment 310. In such cases, the modules 320 may be washable. Additionally or alternatively, in certain embodiments, one or more of the modules 320 may be removable and replaceable from the garment 310. In such cases, the modules 320 need not be washable, although the modules 320 may be designed to be washable and / or to be durable enough to withstand prolonged contact with the designated areas 312 of the garment 310. The modules 320 may be positionable in more than one of the designated areas 312 of the garment 310. That is, one or more of the multiple modules 320 may be configured to sense data using physiological sensors 322 in multiple designated areas 312 of the garment 310.

[0105] The module 320 may include one or more physiological sensors 322 and a communication interface 324 programmed to transmit data from at least one of the physiological sensors 322. For example, the physiological sensors 322 may include one or more of a heart rate monitor (e.g., one or more PPG sensors), an oxygen monitor (e.g., a pulse oximeter), a blood pressure monitor, a thermometer, an accelerometer, a gyroscope, a location sensor, a global positioning system (GPS), a clock, a electrodermal response (GSR) sensor, or other electrical, acoustic, optical, or other sensor or combination of sensors useful for physiological, environmental, or other monitoring described herein. In one embodiment, the physiological sensors 322 may include a conductivity sensor used for electromyography, electrocardiography, electroencephalography, or other physiological sensing based on electrical signals. The data received from the physiological sensors 322 may include at least one of heart rate data and / or similar data related to blood flow (e.g., from a PPG sensor), muscle oxygen saturation data, temperature data, movement data, position / location data, environmental data, time data, blood pressure data, etc.

[0106] Accordingly, certain embodiments include one or more physiological sensors 322 configured to provide continuous heart rate measurements, such as using photoplethysmography. The physiological sensors 322 may include one or more light emitters for emitting light at one or more desired frequencies toward the user's skin and one or more photodetectors for receiving light reflected from the user's skin. The photodetectors may include photoresistors, phototransistors, photodiodes, etc. A processor may process the optical data from the photodetectors to calculate the heart rate based on the measured reflected light. To minimize or eliminate noise in the heart rate signal caused by movement or other artifacts, the optical data may be combined with data from one or more motion sensors, such as an accelerometer or gyroscope. The physiological sensors 322 may also or alternatively perform at least one of continuous motion detection, environmental temperature detection, electrodermal activity (EDA), and / or galvanic skin response (GSR).

[0107] The system 300 may include different types of modules 320. For example, multiple different modules 320 may each provide a specific function. Thus, the garment 310 may house one or more of a temperature module, a heart rate / PPG module, a muscle oxygen saturation module, a tactile module, a wireless communication module, or a combination thereof, any of which may be integrated into a single module 320 or located in separate modules 320 that can communicate with each other. Some measurements, such as temperature, motion, and optical heart rate detection, may have preferred or fixed locations, and pockets or fixtures within the garment 310 may be adapted to house certain types of modules 320 in specific locations within the garment 310. For example, motion may be preferentially detected at or near the extremities, and heart rate data may be preferentially collected near major arteries. In another embodiment, some measurements, such as temperature, can be measured anywhere but may be preferred to be measured in a single location to avoid certain calibration issues that may arise from free placement.

[0108] In another embodiment, the system 300 may include two or more modules 320 located at different locations and configured to perform differential signal analysis. For example, pulse propagation velocity and attenuation in a heartbeat signal may be detected at two or more locations using two or more modules, such as at the user's biceps and wrist, or at other similarly located locations along the arteries. These multiple measurements support differential analysis, which enables useful inferences about heart strength, circulatory flexibility, blood pressure, and other aspects of the cardiovascular system that may indicate cardiac age, cardiac health, cardiac condition, and the like. Similarly, muscle activity detection may be measured at different locations to enable differential analysis to identify activity types or determine muscle fitness. More generally, differential analysis is facilitated by multiple sensors. To perform this type of analysis more accurately, the garment infrastructure may include a beacon or clock to synchronize signals between multiple modules. This is particularly true when data is temporarily stored locally in each module or transmitted wirelessly from different locations to a processor (where real-time processing is difficult due to packet loss, latency, etc.).

[0109] The communication interface 324 may be any of those described herein, and may include, for example, the functionality of any of the network interfaces 304 described above.

[0110] The controller 330 may be configured to determine the location of the module 320, for example, via computer-executable code. This may be based on contextual measurements, such as accelerometer data, from the module 320, which may be analyzed, for example, via machine learning models, to infer body position. In another embodiment, this may be based on other signals from the module 320. For example, signals from sensors such as photodiodes, temperature sensors, resistors, capacitors, etc., may be used alone or in combination to infer body position. In another embodiment, the position may be determined based on the distance from the module 320 to a proximity sensor, RFID tag, etc., located at or near one of the designated areas 312 of the garment 310. Based on the position, the controller 330 may modify the operation of the module 320 to be location-specific. This may include selecting filters, processing models, physiological signal detection, etc. It will be appreciated that the operations of controller 330 (which may be any controller, microcontroller, microprocessor, or other processing circuitry, etc.) may be performed in cooperation with other components of system 300, such as processor 340, one or more of modules 320, or another computing device described herein. It will also be appreciated that controller 330 may be located in a local component of system 300 (e.g., on garment 310, in module 320, etc.), as part of remote processing facility 350, or a combination thereof. Thus, in one embodiment, controller 330 may be included in at least one of multiple modules 320. In another embodiment, controller 330 may be a component separate from garment 310 and serve to coordinate the functionality of various modules 320 connected to garment 310. Controller 330 may also or instead be located remotely from each of multiple modules 320, or some combination thereof.

[0111] Controller 330 may be configured to control one or more of: (i) the sensing performed by physiological sensors 322 of module 320; and (ii) the processing by module 320 of data received from physiological sensors 322. That is, in certain embodiments, the combination of sensors in module 320 may differ based on where it is positioned on garment 310. In other embodiments, the processing of data from module 320 may differ based on where it is positioned on garment 310. In the latter embodiment, a processing resource such as controller 330 or other local or remote processing resource coupled to module 320 can detect location and adapt the processing of data from module 320 depending on the location. This may include, for example, selecting different models, algorithms, or parameters for processing the sensed data.

[0112] In another aspect, this may involve selecting among a variety of different activity recognition models based on the detected location. For example, a variety of different activity recognition models, such as machine learning models, lookup tables, analytical models, etc., may be created and applied to the accelerometer data to detect the type of activity. Other motion data, such as gyroscope data, may also or instead be used. The activity recognition process may also be augmented with other potentially relevant data, such as data from a barometer, magnetometer, GPS system, etc. This may generally distinguish between, for example, sleeping, resting, or exercising, or may more finely distinguish between different types of athletic activities, such as walking, running, cycling, swimming, tennis, squash, etc. While a useful model for detecting activity may be created in this way, the nature of the detection depends on where the accelerometer is located on the body. Therefore, it is useful for processing resources to first identify location using a location detection system (e.g., tag, electromechanical bus connection, etc.) integrated into the garment 310 and then use this detected location to select an appropriate model for activity recognition. This technique may be applied to calibration models, physiological signal processing models, etc. as well, or may adapt the processing of signals from module 320 depending on the location of module 320.

[0113] Identifying the location of the module 320 may include receiving a sensed location of the module 320. The sensed location may be obtained from proximity detection circuitry, such as a near-field communication (NFC) tag, an RFID tag (active or passive), a capacitance sensor, a magnetic sensor, electrical contacts, and mechanical contacts. Any such proximity detection-enabled hardware may be located in the module 320 and the garment 310, and communication may occur between them as needed to detect location. For example, in one embodiment, an NFC tag may be located on or in the garment 310, and the module may include an NFC tag sensor 320 capable of detecting the tag and reading location-specific information from the tag. Proximity detection may additionally or alternatively be performed using capacitive contact detection, electromagnetic proximity detection, mechanical contact, electrical coupling, etc. In this manner, the garment 310 may provide information to the attached module 320 to, among other things, inform the module 320 of where the module 320 is located, or vice versa.

[0114] In this manner, communication between the module 320 and the garment 310 (or a processor in the garment 310) can be used to determine the location of the module 320 on the garment 310. Communication of location information can be accomplished using active or passive techniques, or a combination thereof. For example, thin, flexible, inexpensive, and washable NFC tags can be sewn onto the garment 310 in various locations where the module 320 can be placed. Once the module 320 is placed on the garment 310, the module 320 can interrogate adjacent NFC tags to identify its location. Additionally, other information may be provided to the module 320 via NFC or other similar technologies. This may include detailed information about the garment 310, such as size, gender-specificity, manufacturer information, garment model and serial number, or stock keeping unit (SKU). Similarly, the tag may be encoded with a unique identifier for the garment 310, which can be used to retrieve other relevant information using online resources. The module 320 may also or instead communicate information about itself to the garment 310. This allows the garment 310 to synchronize processing with other modules 320, synchronize communication between modules 320, and control or adjust signals from modules 320. Modules 320 can then configure themselves depending on the current situation of the garment 310 and associated modules 320 and / or perform certain types of monitoring or data processing.

[0115] Determining the location of module 320 may additionally or alternatively be based at least in part on interpretation of data received from physiological sensors 322 of module 320. For example, movement of module 320 detected by the sensors may provide information that can be used to predict location on or within garment 310. Additionally or alternatively, the type of data received from module 320 may indicate where module 320 is located on garment 310. For example, where module 320 is located may generate a unique signature, such as acceleration, gyroscope activity, capacitance data, optical data, temperature data, etc. This data may be fused and analyzed in any suitable manner to predict location.

[0116] As mentioned above, determining the location of the module 320 may additionally or alternatively include receiving explicit input from the user 301. This input may identify one of the designated areas on the garment 310 or a general area of ​​the body (e.g., left wrist, right ankle, etc.). Because the location of the module 320 relative to the garment 310 may be determined from an analysis of multiple data sources, the system 300 may include a component (e.g., the processor 340) configured to adjust one or more potential sources of location information based on expected reliability, the quality of the measured data, explicit user input, etc. In this situation, a confidence level of the prediction may also be advantageously generated. The confidence level of the prediction may be used, for example, to determine whether to query the user for more specific location information. More generally, if the location cannot be reliably predicted, any of the above techniques may be used in conjunction with or in combination with a fail-safe measure that requires user input. Additionally or alternatively, the user may explicitly specify a prediction in advance, i.e., override an automatically generated prediction.

[0117] The location of module 320 determined using any of the above techniques may be transmitted to a remote processing facility 350, database 360, or the like, for storage and analysis. That is, module 320 may use this information locally to configure itself to the location of module 320, but may also communicate this information to other modules 320, peripheral devices, or the cloud. Processing this information in the cloud may help an organization determine whether module 320 has been attached to garment 310, which locations are most used, and how module 320 performs differently in different locations. These analyses may serve many purposes, including, for example, to improve the design or use of module 320 and garment 310 for a population, a type of user, or a specific user.

[0118] As described above, the system 300 may further include a processor 340 and a memory 342. Generally, the memory 342 may hold computer-executable code configured to be executed by the processor 340 to perform processing of data received from one or more of the modules 320. One or more of the processor 340 and the memory 342 may be located on a local component of the system 300 (e.g., the garment 310, the modules 320, or the controller 330, etc.) or may be located as part of a remote processing facility 350, etc., as shown. Thus, in one embodiment, the processor 340 and one or more of the memory 342 may be included in at least one of the multiple modules 320. In this manner, processing may be performed in a central module or independently in each module 320. In another embodiment, the processor 340 and one or more of the memory 342 may be located remotely with respect to each of the multiple modules 320. For example, processing may be performed in a connected peripheral device, such as a smartphone, laptop, local computer, or cloud resources.

[0119] The memory 342 may store one or more algorithms, models, and supporting data (e.g., parameters, calibration results, user preferences, etc.) for transforming data received from the physiological sensors 322 of the module 320. In this manner, appropriate models, algorithms, tuning parameters, etc. for use in transforming data may be selected based on the location of the module 320 determined by the controller 330 and / or processor 340, as described herein. For example, an algorithm for transforming data from the accelerometer of the module 320 into activity data or a user's step count may differ depending on whether the module 320 is worn on the user's wrist or on the user's waistband. Similarly, the intensity of an LED and the sensitivity of its corresponding photodetector may differ when the PPG device is worn on the wrist versus the thigh. Thus, the module 320 may configure itself for the location by controlling one or more of the sensor type, sensor parameters, processing model, etc., based on the detected location of the module 320.

[0120] The selection of the algorithm may additionally or instead include an analysis of one or more of sensor data, metadata, and the like. For example, the algorithm may be selected at least in part based on metadata received from one of the module 320 and the garment 310. This metadata may be derived from communication between the module 320 and the garment 310, such as from an exchange of information between a tag and a tag reader. For example, the garment 310 may include garment-specific metadata stored on a tag or other wirelessly readable data source, such as an NFC tag, which may be readable by or transmittable to one or more of the modules 320, the controller 330, and the processor 340. Such garment-specific metadata may include at least one of the type of garment 310, the size of the garment 310, garment dimensions, gender preference of the garment 310, manufacturer, model number, serial number, SKU, materials, fit information, and the like. In one embodiment, this information may be provided by one or more of the location identification tags described herein. In another embodiment, the garment 310 may include additional tags in suitable locations (such as near or accessible to the processor or controller) that provide garment-specific information, while other tags may provide location-specific information.

[0121] The metadata may further or alternatively include at least one of the user's 301 gender, the user's 301 weight, the user's 301 height, the user's 301 age, metadata related to the garment 310 (e.g., garment size, type, material, etc.), etc. The metadata may be derived, at least in part, from input information provided by the user or may be derived from information derived from the user 301 (e.g., the user's account information as a participant in the system 300, etc.). For example, a processing algorithm may be selected depending on the garment's 310 material as conveyed by the garment's 310 serial number or model number, the user's 301 physiological characteristics as implied by the garment's size, etc. The metadata may further or alternatively be used to verify the authenticity of the garment 310 or to control access to the garment 310 and / or a module 320 coupled to the garment 310. In one embodiment, the metadata (e.g., size, material, etc.) may be directly encoded in the garment's metadata. In another embodiment, the garment 310 may expose a unique identifier that can be used to obtain relevant information from the manufacturer or other data sources. This latter approach advantageously allows for correlation of garment-specific data with other user-specific data such as height, weight, and body composition.

[0122] Knowing in advance where module 320 is located may allow the use of algorithms tailored to work optimally in that particular location. This can reduce the significant computational burden imposed on module 320 to analytically assess location based on available signals. Other information may also or instead be used to select the optimal algorithm. For example, an algorithm may use different models or different model parameters based on the gender or dimensions of the garment.

[0123] The processor 340 may be configured to evaluate the quality of the data received from the physiological sensor 322 of the module 320. For example, the processor 340 may be configured to provide recommendations regarding at least one of the position of the module 320 and aspects of the garment 310 (e.g., size, fit, material, etc.) based on the quality of the data. For example, the processor 340 may be configured to detect that the garment is not properly fitting the wearer when acquiring physiological data. This may be done, for example, by detecting a situation where the module is moving but the sensed physiological signal data is of low quality or absent (e.g., from accelerometer data). Generally, the garment 310 may store its own identifier and / or metadata, for example, as described herein, or the garment identification data may be stored on a tag (e.g., in a designated area 312 of the garment 310). The processor 340 may be configured to use this garment identification and / or metadata to provide the user 301 with recommendations regarding alternative garments 310 or adjustments to the current garment 310. For example, if a particular garment 310 is likely to generate poor quality data, the user 301 may be prompted to select a different size or make other adjustments. Additionally, data regarding the number of uses of the garment 301 may be collected to aid in business decisions. This may help determine, for example, which garments 301 generate the highest quality data and which garments 310 are most preferred by the user 301.

[0124] System 300 may further include database 360. Database 360 ​​may be located remotely and in communication with system 300 via data network 302. Database 360 ​​may store data related to system 300 as described herein, such as sensed data, processed data, transformed data, metadata, physiological signal processing models and algorithms, individual activity history, etc. System 300 may further include one or more servers 370 that host data, provide a user interface, or process data to facilitate use of modules 320 and garment 310 as described herein.

[0125] As will be appreciated, the garment 310, module 320 and associated garment infrastructure can be advantageously used in combination with remote network / processing resources to improve physiological monitoring and enable monitoring aspects not previously available.

[0126] One or more of the devices and systems described herein may include circuitry for both wireless charging and wireless data transmission. For example, the corresponding circuits may be operable independently of one another and the corresponding antennas may be located in close proximity to one another (e.g., the circuitry for wireless charging and the circuitry for wireless data transmission may be located along substantially the same plane or may include separate coils located relatively close together within the device or system). In such embodiments, one or more measures may be taken to prevent the wireless data transmission process from interfering with the wireless power transmission process. More specifically, measures may be taken by coupling the data circuitry to the electromagnetic field for wireless power transfer in a manner that reduces its efficiency in charging the device by altering its resonant frequency or by destructively interfering with the power transfer. For example, including a switch to disable the circuitry for data transmission when a specific wireless charging activity is present allows for efficient wireless charging of the device relatively unhindered. The switch may also be operable to enable operation of the data transmission circuitry when a specific wireless charging activity is not present.

[0127] Thus, for example, in the context of the physiological monitor described herein, the physiological monitor may include both a wireless power receiver (or the like) and a wireless data tag reader (or the like). Generally, these subsystems may conform to one or more Near Field Communication (NFC) specifications for protocols and physical architectures, or any other standard suitable for wireless power and data transmission. The power circuitry may be used, for example, to charge the physiological monitor's battery, allowing the device to be recharged without physically connecting it to a power source. The data circuitry may be used, for example, as a wireless data tag reader, or the like, to read data from a nearby data source, such as an identification tag attached to a user's clothing. Generally, the physiological monitor may include separate circuits (separate coils) for these wireless power and data systems, and may include, for example, separate processing circuits and / or separate antennas. The antennas may be positioned along substantially the same plane of the physiological monitor (e.g., one coil may be positioned substantially inside the other coil or adjacent to each other). In one embodiment, the antennas may be arranged in parallel planes; however, because NFC-standard devices have relatively small distance tolerances, it is preferable that the physical housing of these antennas enforce the same or substantially the same distance for both antennas in such an architecture. In this context, the antennas may be positioned as nearly parallel as possible within reasonable manufacturing tolerances. That is, the antennas may be positioned as nearly parallel as possible when located on two different layers of a shared printed circuit board, or preferably when located on a single layer of a shared printed circuit board. The physiological monitor may further include a switch (e.g., a radio frequency (RF) switch) in series with the coil for the wireless data tag reader, which can disable the wireless data tag reader when powered, thereby reducing its impact on the efficiency of the wireless power transfer process.In particular, the switch may be configured to open when powered and to close when the physiological monitor is looking for a data tag to read.

[0128] 4 is a block diagram of a computing device 400. The computing device 400 may be, for example, a device used for continuous physiological monitoring or any other device that supports physiological monitoring in the systems and methods described herein. Additionally or alternatively, the device may be any of the local computing devices described herein, such as a desktop computer, laptop computer, or smartphone. Additionally or alternatively, the device may be any of the remote computing resources described herein, such as a web server, cloud database, file server, application server, or any other remote resource. While described as a physical device, it will be understood that the exemplary computing device 400 may also or alternatively be implemented as a virtual computing device, such as a virtual machine running a web server or other remote resource on a cloud computing platform. Generally, the device 400 may include one or more sensors 402, a battery 404, storage 406, a processor 408, memory 410, a network interface 414, and a user interface 416, or one or more virtual instances of the foregoing.

[0129] The sensors 402 may include any sensor or combination of sensors suitable for heart rate monitoring as contemplated herein, and sensors 402 for detecting calorie expenditure, location (e.g., via a global positioning system, etc.), movement, activity, etc. In one embodiment, this may include a light sensing system including an LED or other light source and a photodiode or other light sensor. The light source and light sensor may be used in combination to provide photoplethysmography of heart rate, pulse oximetry, and other physiological monitoring.

[0130] Sensor 402 may additionally or alternatively include one or more sensors for activity measurement. In some embodiments, the system may include one or more multi-axis accelerometers and / or gyroscopes to measure activity. In some embodiments, the accelerometer may also be used to filter signals from an optical sensor for measuring heart rate to obtain more accurate heart rate measurements. In some embodiments, the wearable system may include a multi-axis accelerometer to measure movement and calculate distance. A motion sensor may be used to classify or categorize activities such as walking, running, playing another sport, standing, sitting, or lying down. For example, sensor 402 may include a thermometer to monitor the user's body temperature or skin temperature. In one embodiment, sensor 402 may be used to recognize sleep based on a decrease in temperature, electrodermal response data, a lack of movement or activity according to data collected by an accelerometer, a decrease in heart rate measured by a heart rate monitor, etc. Body temperature, in combination with heart rate monitoring and movement, may be used to interpret, for example, whether a user is asleep or simply resting, and how well an individual is sleeping. The temperature, movement, and other sensed data may also be used to identify whether the user is exercising and to classify and / or analyze the activity, as described in more detail below. In another embodiment, the sensors 402 may include one or more contact sensors, such as capacitive or resistive touch sensors, for detecting the placement of a physiological monitor relative to the user. More generally, the sensors 402 may include any sensor or combination of sensors suitable for monitoring geographic location, physiological status, movement, motion, and the like, in any manner useful for physiological monitoring as contemplated herein.

[0131] Battery 404 may include one or more batteries configured to allow for continuous wearing and use of the wearable system. In one embodiment, the wearable system may include two or more batteries, such as a removable battery that can be removed and recharged using a charger and an integrated battery that maintains operation of device 400 while the main battery is charging. In another aspect, battery 404 may include a wirelessly rechargeable battery that can be recharged using a short-range or long-range wireless charging system.

[0132] The processor 408 may include any microprocessor, microcontroller, signal processor, or other processor, or combination of a processor and other processing circuitry, suitable for performing the processing steps described herein. Generally, the processor 408 is configured by computer-executable code stored in the memory 410 to provide the activity recognition and other physiological monitoring functions described herein.

[0133] Generally, memory 410 may include one or more non-transitory computer-readable media for storing one or more computer-executable instructions or software for implementing the exemplary embodiments. Non-transitory computer-readable media may include, but are not limited to, one or more types of hardware memory, non-transitory tangible media (e.g., one or more magnetic storage disks, optical disks, USB flash drives), etc. In one aspect, memory 410 may also include computer system memory or random access memory such as DRAM, SRAM, EDORAM, etc. Memory 410 may also include other types of memory, or combinations thereof, and virtual instances of memory (e.g., where the device is a virtual device). Generally, memory 410 may store computer-readable and computer-executable instructions or software for implementing the methods and systems described herein. Memory 410 may also or alternatively store physiological data, user data, or other data useful for the operation of the physiological monitor or other device described herein (e.g., data collected by sensor 402 during operation of device 400).

[0134] The network interface 414 may be configured to communicate data wirelessly to a server 420, for example, via an external network 418 (e.g., any public, private, or other data network described herein) or via any combination of the foregoing (e.g., including a local area network, the Internet, a cellular data network, etc.). If the device is a physiological monitoring device, the network interface 414 may be used, for example, to transmit raw or processed sensor data stored on the device 400 to the server 420, and may also be used to receive updates, receive configuration information, and communicate with remote resources and users to support device operation. More generally, the network interface 414 may include any interface configured to connect with one or more networks via various connections. The network may be, for example, a local area network (LAN), a wide area network (WAN), the Internet, a cellular data network, or the like. The connections may include, but are not limited to, standard telephone lines, LAN or WAN links (e.g., 802.11, T1, T3, 56kb, X.25), broadband connections (e.g., ISDN, Frame Relay, ATM), wireless connections, or a combination of any or all of the foregoing. Network interface 412 may include a built-in network adapter, a network interface card, a PCMCIA network card, a card bus network adapter, a wireless network adapter, a USB network adapter, or a modem, or any other device suitable for connecting computing device 400 to any type of network having communications capabilities and capable of performing the operations described herein.

[0135] User interface 416 may include any component suitable for supporting interaction with a user. This may include, for example, a keypad, a display, a buzzer, a speaker, a light-emitting diode, a capacitive touch sensor or pad, and any other component for receiving input from or providing output to a user. In one embodiment, device 400 may be configured to receive tactile input, for example, by changing operational states or displaying information in response to a series of tapping motions on the surface of the device. User interface 416 may also or alternatively include a graphical user interface displayed on a display device for graphical interaction with programs executing on processor 408 or other content displayed by a physical display of device 400.

[0136] Techniques for observing, monitoring, analyzing stress, and / or providing coaching or other recommendations regarding stress are described below. By way of background and context, the term stress, as used herein, may include any measurable physiological and / or psychological response to a stimulus. For example, psychological stress may be caused by stimuli such as work pressure, financial difficulties, relationship problems, health concerns, etc. At the same time, physiological stress may be caused by physical demands, sleep cycles, etc. The observed stress response may be caused by either of these causes (physiological or psychological) or a combination of both.

[0137] In one aspect, it may be useful to distinguish between physiological stress caused by physical activity and psychological stress caused by psychological stressors. For example, when a threat is perceived, the brain sends a signal to the hypothalamus, which activates the sympathetic nervous system. This releases hormones such as adrenaline and cortisol, increasing heart rate, respiratory rate, and blood pressure. In small doses, this type of stress can be helpful because it motivates a person to take action and deal with difficult situations. However, when this type of stress is experienced chronically or excessively, it can have a negative impact on a person's mental and physical health, leading to anxiety, depression, fatigue, and weakened immunity. At the same time, physiological stress responses can be a healthy response to strenuous activity. Nevertheless, incremental monitoring of physiological stress responses can support intraday updates, such as performance metrics and coaching recommendations. For example, this can support updating strain calculations and providing real-time recommendations for new or ongoing exercise or dietary changes.

[0138] As an important advantage, the techniques described herein facilitate real-time or near-real-time monitoring and management of psychological and physiological stressors. Another advantage is that the techniques described herein facilitate the separate measurement of both physical and emotional stress responses, allowing, for example, the separate analysis of the effects of physical and emotional stimuli on current stress states.

[0139] 5 illustrates a system for dynamic monitoring. Generally, physiological monitoring system 500 may include a wearable device 502, such as any of the physiological monitors described herein, a remote resource 504, such as any of the servers or other remote processing resources described herein, and a user device 506, such as any of the user devices described herein, as well as a data network 508 interconnecting these devices in a communicative relationship.

[0140] The wearable device 502 can continuously monitor, measure, and / or calculate physiological parameters such as heart rate, heart rate variability, temperature, electrical characteristics (e.g., electrodermal activity), blood pressure, stress, and movement, and transmit this data to the remote resource 504 via a data network 508. Based on this data, the remote resource can calculate metrics such as a daily sleep score (assessing the previous night's sleep), a daily stress score (assessing the previous day's stress), or a daily recovery score (assessing current readiness for new stress), using, for example, techniques described, by way of non-limiting example, in U.S. Pat. No. 10,264,982. The entire contents of U.S. Pat. No. 10,264,982 are incorporated herein by reference. This server-based approach is advantageous in that it offloads data- and computation-intensive processing to a remote server or other appropriately capable computing system, such as when metrics are calculated based on heart rate and movement data over an entire 24-hour period. However, this approach is less effective for incrementally updating data and recommendations throughout the day.

[0141] Thus, the techniques described herein can be advantageously deployed to dynamically monitor activity throughout the day and update the user's metrics and coaching recommendations as needed. More specifically, the dynamic monitor 510 may be located locally on the wearable device 502 or in another convenient location (e.g., the user device 506) where local calculations can be performed frequently to dynamically update user information. This approach is also advantageous in that it allows for rapid detection of significant stress events and the recommendation of appropriate interventions. In this context, "dynamic" scoring is understood to refer to scoring performed incrementally between static remote calculations over longer time periods, such as a day or even several hours. While this dynamic scoring is necessarily performed over discrete time periods, it may be updated in any periodic or substantially continuous manner so that the user receives a timely quantitative assessment. This may include, for example, updates that are as close to instantaneous as possible, so that the user experiences updates in real time with little or no perceptible delay. In another aspect, where the calculations are more complex and / or processed remotely, the current score of any indicator may be calculated and updated to the user at intervals such as once per minute, or at shorter or longer intervals suitable for the user's use. This may include changing the frequency, for example, increasing the frequency of updates while the user is viewing the stress score and / or providing more up-to-date calculations. More generally, any amount or frequency of updates that allows for dynamic tracking of the user and / or supports feedback to the user regarding their current state may be used to support tracking and reporting of the user's current state of subjective or physiological stress.

[0142] FIG. 6 is a flow diagram illustrating an exemplary signal processing algorithm for generating a heart rate sequence for each detected heartbeat. The signal processing algorithm is embodied in computer-executable instructions stored on one or more non-transitory computer-readable media. In step 602, a light emitter of the wearable physiological measurement system emits light toward the user's skin. In step 604, light reflected from the user's skin is detected by a photodetector within the system. In step 606, signals or data associated with the reflected light are preprocessed using any suitable technique to enable heartbeat detection. In step 608, a processing module of the system executes one or more computer-executable instructions associated with a peak detection algorithm to process the data corresponding to the reflected light to detect multiple peaks associated with multiple beats of the user's heart. In step 610, the processing module identifies RR intervals based on the multiple peaks detected by the peak detection algorithm. In step 612, the processing module determines a confidence level associated with the RR intervals.

[0143] Based on the confidence associated with the RR interval estimate, the processing module selects either a peak detection algorithm or a frequency analysis algorithm to process the data corresponding to the reflected light to determine the user's instantaneous heart rate sequence. The frequency analysis algorithm can process the data corresponding to the reflected light based on the user's movement detected using, for example, an accelerometer. The processing module can select the peak detection algorithm or the frequency analysis algorithm regardless of the user's state of movement. Because certain frequency-based approaches cannot obtain accurate RR intervals for heart rate variability analysis, it is advantageous to use the estimated confidence to determine whether to switch to a frequency-based method. Therefore, in one embodiment, the information that can be extracted is maximized by maintaining the ability to obtain RR intervals for as long a time as possible, even in the presence of movement.

[0144] For example, step 614 determines whether the confidence associated with the RR interval exceeds (or is equal to or greater than) a threshold. In certain embodiments, the threshold may be preset, e.g., approximately 50%-90% in some embodiments, and approximately 80% in one non-limiting embodiment. In other embodiments, the threshold may be context-dependent. That is, the threshold may be dynamically and automatically determined based on past confidence levels. For example, if one or more past confidence levels were high (i.e., above a particular level), the system may determine that a relatively low current confidence level compared to the past levels indicates a low-confidence signal. In this case, the threshold may be dynamically adjusted to a higher value, and a frequency-based analysis method may be selected to process the low-confidence signal.

[0145] If the confidence is greater than (or equal to or greater than) the threshold, the processing module may use the multiple peaks to identify the user's instantaneous heart rate in step 616. On the other hand, in step 620, based on a determination that the confidence associated with the RR interval is less than or equal to a predetermined threshold, the processing module may execute one or more computer-executable instructions associated with a frequency analysis algorithm to identify the user's instantaneous heart rate. The confidence threshold may be dynamically set based on past confidence levels.

[0146] In some embodiments, at step 618 or 622, the processing module may determine the user's heart rate variability based on the instantaneous heart rate / beat sequence.

[0147] The system may include a display device configured to display a user interface for displaying the sequence of instantaneous heart rates, RR intervals, and / or heart rate variability of the user identified by the processing module. The system may include a storage device configured to store the sequence of instantaneous heart rates, RR intervals, and / or heart rate variability identified by the processing module.

[0148] In one embodiment, the system can switch between different analysis techniques for determining heart rate, such as a statistical technique for detecting heart rate and a frequency-domain technique for detecting heart rate. These two different modes have different advantages in terms of accuracy, processing efficiency, and information content, and therefore may be useful at different times and under different conditions. Rather than selecting one of such modes or techniques in an attempt to optimize, the system may conveniently switch between these different techniques or other analysis techniques using predetermined criteria. For example, if a statistical technique is used, a confidence level may be determined and used as a threshold for switching to an alternative technique, such as a frequency-domain technique. Additionally or alternatively, the threshold may be determined in response to historical, subjective, and / or adaptive data for a particular user. For example, the selection of the threshold may be determined in response to data for a particular user, including, but not limited to, subjective information about how a particular user's heart rate changes in response to stress, exercise, etc. Similarly, the threshold may be changed in response to changes in the user's fitness, context provided by other sensors in the wearable system, signal noise, etc.

[0149] An exemplary statistical technique employs probabilistic peak detection. In this technique, discrete probability steps are established, and a likelihood function may be established as a mixture of Gaussian random variables and uniform distributions. The core of the likelihood function encodes the assumption that the peak detection algorithm produced a reasonable initial estimate with a first probability (p), but not with a second probability (1-p). In the next step, Bayes' theorem is applied to determine the posterior density over the parameter space, and its maximum value (i.e., the argument (parameter) that maximizes the posterior distribution) is obtained. This value becomes the heart rate estimate. In the next step, the previous two steps are applied again to the remaining samples. Due to processing noise, some signal variation occurs, which depends on the length of the interval. This processing noise manifests itself as the variance of the Gaussian distribution used in the likelihood function. Based on the new posterior distribution, an estimate such as the maximum posterior probability is then determined. A confidence value for the estimate is recorded, and the posterior values ​​at various points in the parameter space, centered around the estimate + / - precision, are summed to provide a precision measure.

[0150] The beats per minute (BPM) parameter space θ may range from about 20 to about 240, corresponding to the empirical limits of human heart rate. In an exemplary method, a probability distribution is computed over this parameter space, and at each step, the mode of the distribution is declared as the heart rate estimate. A discrete uniform prior can be set as follows:

[0151]

number

[0152] The univariate likelihood before normalization is defined by a mixture of Gaussian and uniform functions as follows:

[0153]

number

[0154] however,

[0155]

number

[0156] where σ and p are predetermined constants.

[0157] Applying Bayes' theorem, the posterior density with respect to θ is determined, which can be expressed as, for example, the prior density vector

number

number

number

[0158]

number

[0159] If k≧2, the variance of the signal S(t) due to processing noise is determined. Next, the following variables are set to give more process / peak-to-peak noise for longer RR intervals, and the normalized convolution is set as follows:

[0160]

number

[0161] where f is the density function below.

[0162]

number

[0163] Next, calculate the following formula:

[0164]

number

[0165] This formula is then normalized and recorded as:

[0166]

number

[0167] Finally, the confidence level of the above formula for a particular accuracy threshold is determined as follows:

[0168]

number

[0169] An exemplary frequency analysis algorithm used in one embodiment may isolate the highest frequency component of the optical data, check for harmonics common to both the accelerometer data and the optical data, and perform filtering of the optical data. The algorithm takes as input the raw analog signals from the accelerometer (3-axis) and pulse sensor, and outputs a value for heart rate, or beats per minute (BPM), within a given time period associated with the spectrogram window.

[0170] The isolation of the highest frequency component is performed in multiple stages, gradually narrowing the window size under consideration, thereby reducing the margin of error. In one embodiment, a spectrogram of 2 samples may be generated, including overlap with 2 samples of optical data. The spectrogram is limited to frequencies where a heart rate may be present. When the window size under consideration is reduced, these limiting boundaries may be updated. A frequency estimate is extracted from the spectrogram of the optical data by identifying the most prominent frequency component of the spectrogram. The frequency may be extracted using the following exemplary steps: Identify the most prominent frequency of the spectrogram on the signal; Determine whether the frequency estimate is a harmonic of the true frequency; If the frequency estimate is a harmonic of the true frequency, replace the frequency estimate with the true frequency; Determine whether the current frequency estimate is a harmonic of the motion sensor data; If the frequency estimate is a harmonic of the motion sensor data, replace the frequency estimate with a past time estimate; Save the upper and lower bounds of the obtained frequency; Optionally, add or subtract a constant value. Subsequent steps may narrow the search range by adding or subtracting fewer constants. Some of the previous steps are repeated one or more times (e.g., three times), except for taking 2(15-i) samples for the window size and 2(13-i) samples if there are overlaps in the spectrogram, where i is the current iteration number. The final output is the final symmetric endpoint average of the frequency estimates.

[0171] The following table shows the performance of the algorithms disclosed herein. To obtain the results below, experiments were conducted in which subjects wore an exemplary wearable physiological measurement system and a three-lead ECG, both wired to the same microcontroller (e.g., Arduino) to provide time-synchronized data. Approximately 50 data sets were analyzed, including data while the subjects were stationary, walking, and running on a treadmill. [Table 1]

[0172] The algorithm's performance stems from a combination of probabilistic and frequency-based approaches. Three challenges in creating an algorithm to calculate heart rate from PPG data are: 1) false beat detection, 2) missing actual beat detection, and 3) errors in the precise timing of beat detection. The algorithm disclosed herein improves upon these three sources of error, in some cases consistently achieving within 2 BPM of the ECG value, even during the most intense exercise.

[0173] An exemplary wearable system calculates heart rate variability (HRV) to understand the body's state of recovery. These values ​​are obtained either just before the user wakes up or when the user is motionless. In either case, the variability of a photoplethysmography (PPG) is comparable to the HRV of an electrocardiogram (ECG). HRV is traditionally measured by capturing a time series of R-R intervals using an ECG machine. Because the exemplary wearable system uses PPG, instead of capturing electrical signatures from the heartbeat, the peaks in the captured signal correspond to arterial blood volume. At rest, these peaks directly correlate with the cardiac cycle, and HRV can be calculated by analyzing the peak-to-peak intervals (the intervals in PPG that are analogous to the R-R intervals). Medical literature has demonstrated that these peak-to-peak intervals, or "PPG variability," are identical to resting HRV in an ECG. See the following references: Charlot K et al., "Interchangeability between heart rate and photoplethysmography variabilities during sympathetic stimulation," Physological Measurement, 2009 December;30(12):1357-69. doi:10.1088 / 0967-3334 / 30 / 12 / 005, URL: http: / / www.ncbi.nlm.nih.gov / pubmed / 19864707; and Lu, S et al., "Can photoplethysmography variability serve as an alternative approach to obtain heart rate variability information?", Journal of Clinical Monitoring and Computing, 2008 Feb;22(1):23-9, URL: http: / / www.ncbi.nlm.nih.gov / pubmed / 17987395, the entire contents of which are incorporated herein by reference.

[0174] The exemplary physiological measurement system is configured to minimize power consumption, allowing the system to be worn continuously without the need for frequent recharging. The majority of current consumption in the exemplary system is allocated to powering light emitters, such as LEDs, the wireless transceiver, the microcontroller, and peripherals. In one embodiment, the system's circuit board may include a boost converter that drives approximately 10 mA of current to each light emitter at approximately 80% efficiency, allowing the system to draw power directly from the battery at a substantially constant power. For an exemplary battery of approximately 3.7 V, the current draw from the battery may be approximately 40 mW. In some embodiments, the wireless transceiver may draw approximately 10-20 mA of current when actively transferring data. In some embodiments, the microcontroller and peripherals may draw approximately 5 mA of current.

[0175] An exemplary system may include a processing module configured to automatically adjust operating characteristics of one or more of the light emitters and / or light detectors to minimize power consumption while ensuring reliable and continuous detection of all of the user's heartbeats, including, but not limited to, the frequency of light emitted by the light emitters, the number of active light emitters, the duty cycle of the light emitters, the brightness of the light emitters, the sampling rate of the light detectors, etc.

[0176] The processing module may adjust operational characteristics based on one or more signals or indicators obtained or derived from one or more sensors in the system, which may include, but are not limited to, a user's operational state, a user's sleep state, historical information regarding the user's physiology and / or habits, environmental or situational conditions (e.g., ambient light conditions), a user's physical characteristics (e.g., optical characteristics of the user's skin), etc.

[0177] In one embodiment, the processing module may receive data regarding a user's motion, for example, using an accelerometer. The processing module processes the motion data to identify a user motion state indicating the user's motion level (e.g., exercise, light motion (e.g., walking), no motion or rest, sleep, etc.). The processing module may adjust the duty cycle of one or more light emitters and the corresponding sampling rate of one or more photodetectors based on the user's motion state. For example, if the motion state is determined to indicate that the user is at a first high motion level, the processing module may activate the light emitters at a first high duty cycle and sample the reflected light using a photodetector sampling at a first high sampling rate. If the motion state is determined to indicate that the user is at a second low motion level, the processing module may activate the light emitters at a second low duty cycle and sample the reflected light using a photodetector sampling at a second low sampling rate. That is, the duty cycle of the light emitters and the corresponding sampling rate of the photodetectors may be adjusted in a stepwise or continuous manner based on the user's motion state or level. This adjustment ensures that heart rate data is detected frequently enough during movement to ensure that all of the user's heartbeats are detected.

[0178] In a non-limiting example, the light emitters may be operated at a duty cycle ranging from about 1% to about 100%. In another example, to minimize power consumption, the light emitters may be operated at a duty cycle ranging from about 20% to about 50%. Specific exemplary sampling rates of the photodetector range from about 50 Hz to about 1000 Hz, but are not limited to these exemplary rates. Specific non-limiting sampling rates include, for example, about 100 Hz, 200 Hz, 500 Hz, etc.

[0179] In one non-limiting example, while a user is performing an exercise routine, the photodetector may continuously sample to maintain a standard deviation of error within 5 beats per minute (BPM). When the user is resting, the photodetector may be activated at approximately a 1% duty cycle, i.e., only 10 milliseconds per second (i.e., only 1% of the time), which may maintain a standard deviation of error within 5 BPM (including a standard deviation of error in heart rate measurement of 2 BPM and a standard deviation of error in heart rate change between measurements of 3 BPM). When the user is engaged in light activity (e.g., walking), the photodetector may be activated at approximately a 10% duty cycle, i.e., only 100 milliseconds per second (i.e., only 10% of the time), which may maintain a standard deviation of error within 6 BPM (including a standard deviation of error in heart rate measurement of 2 BPM and a standard deviation of error in heart rate change between measurements of 4 BPM).

[0180] The processing module can adjust the brightness of one or more light emitters by adjusting the current supplied to the light emitters. For example, a first brightness level may be set by a current in a range from about 1 mA to about 10 mA, but is not limited to this exemplary range. A second brightness level may be set by a current in a range from about 11 mA to about 30 mA, but is not limited to this exemplary range. A third brightness level may be set by a current in a range from about 80 mA to about 120 mA, but is not limited to this exemplary range. In one non-limiting example, the first, second, and third brightness levels may be set by currents of about 5 mA, about 20 mA, and about 100 mA, respectively.

[0181] In some embodiments, the processing module may detect environmental or situational conditions (e.g., ambient light levels) and adjust the brightness of the light emitters accordingly to ensure that the light detectors can detect light reflected from the user's skin while minimizing power consumption. For example, if the ambient light is determined to be at a first high level, the brightness of the light emitters may be set to the first high level. If the ambient light is determined to be at a second low level, the brightness of the light emitters may be set to a second low level. In some cases, the brightness may be adjusted continuously based on the detected environmental conditions.

[0182] In some embodiments, the processing module may detect a physiological condition of the user (e.g., an optical characteristic of the user's skin) and adjust the brightness of the light emitters accordingly to ensure that the light detector can detect light reflected from the user's skin while minimizing power consumption. For example, if the user's skin is determined to be highly reflective, the brightness of the light emitters may be set to a first, low level. If the user's skin is determined to be less reflective, the brightness of the light emitters may be set to a second, high level.

[0183] An LED with a shorter wavelength may require more power than an LED with a longer wavelength. Therefore, to conserve battery life, an exemplary wearable system may provide and use light emitting at two or more different frequencies based on the detected activity level. For example, if the activity state is determined to indicate that the user is at a first high activity level (e.g., exercising), one or more light emitters may be activated to emit light at a first wavelength. If the activity state is determined to indicate that the user is at a second low activity level (e.g., resting), one or more light emitters may be activated to emit light at a second wavelength longer than the first wavelength. If the activity state is determined to indicate that the user is at a third low activity level (e.g., sleeping), one or more light emitters may be activated to emit light at a third wavelength longer than the first and second wavelengths. Other activity levels may be predetermined and corresponding emission wavelengths may be selected. A threshold activity level that triggers an adjustment of the light wavelength may be determined based on one or more factors (e.g., including, but not limited to, skin characteristics, ambient light conditions, etc.). Any suitable combination of light wavelengths may be selected, such as green (for high operating levels) / red (for low operating levels); red (for high operating levels) / infrared (for low operating levels); and blue (for high operating levels) / green (for low operating levels).

[0184] Short wavelength LEDs may require more power than other types of heart rate sensors, such as piezoelectric or infrared sensors. Therefore, an exemplary wearable system may provide and use a unique combination of sensors to conserve battery life, such as one or more photodetectors for periods of expected movement and one or more piezoelectric and / or infrared sensors for periods of low movement (e.g., during sleep). In other specific embodiments of the wearable system, the piezoelectric and / or infrared sensors may be omitted.

[0185] For example, if the motion state is determined to indicate that the user is at a first high level of motion (e.g., exercising), one or more light emitters may be activated to emit light at a first wavelength. If the motion state is determined to indicate that the user is at a second low level of motion (e.g., resting), a non-light sensor may be activated. The threshold level of motion that triggers adjustment of the sensor type may be determined based on one or more factors (e.g., including, but not limited to, skin characteristics, ambient light conditions, etc.).

[0186] The system may determine the type of sensor to use at a given time based on the level of movement (e.g., via an accelerometer) and whether the user is asleep (e.g., based on movement input, skin temperature, and heart rate). Based on a combination of these factors, the system selects the type of sensor to use to monitor the user's heart rate. Common signs of sleep may include periods of inactivity or intermittent small movements (e.g., turning over in bed), a drop in skin temperature (not significantly lower than normal), a large change in GSR (galvanic skin response), and a heart rate below the user's normal resting heart rate when awake. Because these variables depend on a person's physiological function, machine learning algorithms are trained using user-specific input information to determine whether the user is awake or asleep and, based on that, determine the exact parameters under which the algorithm considers someone asleep.

[0187] In an exemplary configuration, the photodetector may be located on the underside of the wearable system, and all heart rate sensors may be located adjacent to one another. For example, the sensors may be selected and located where the strongest signal occurs, so that a low-power sensor(s) may be adjacent to a high-power sensor(s). In one exemplary configuration, a 3-axis accelerometer may be used and may be located on the top of the wearable system.

[0188] In some embodiments, the processing module may be configured to automatically adjust the rate at which data is transmitted by the wireless transmitter to minimize power consumption while ensuring that raw and processed data generated by the system is transmitted to the external computing device. In one embodiment, the processing module may determine the amount of data to be transmitted (e.g., based on the amount of data generated since the last data transmission) and select the next data transmission time based on the amount of data to be transmitted. For example, if the amount of data is determined to exceed (or be equal to or greater than) a threshold level, the processing module may transmit the data or schedule a time to transmit the data. On the other hand, if the amount of data is determined to not exceed (or be equal to or less than) the threshold level, the processing module may postpone the data transmission to minimize transmitter power consumption. In one non-limiting example, the threshold may be set to the amount of data that can be transmitted in two seconds under current conditions. Exemplary data transmission rates range from approximately 50 kilobytes / second to approximately 1 megabyte / second, but are not limited to this exemplary range.

[0189] In some embodiments, the operating characteristics of the microprocessor may be automatically adjusted to minimize power consumption, and this adjustment may be made based on the user's level of physical activity.

[0190] More generally, the above description contemplates various techniques for directly sensing (e.g., as the reliability or accuracy of a calculated heart rate) or indirectly sensing (e.g., from motion detection or temperature) conditions related to heart rate monitoring or related physiological activity. However measured, however, these sensed conditions can be used to intelligently select among several different modes (e.g., including hardware modes, software modes, and combinations of the foregoing) for monitoring heart rate based on, for example, accuracy, power usage, and detected activity state. Accordingly, techniques are disclosed herein for selecting among two or more different heart rate monitoring modes in response to sensed conditions.

[0191] II. Exemplary Physiological Analysis Systems

[0192] Exemplary embodiments provide an analysis system for qualitative and quantitative monitoring of a user's physique, health, and physical training. The analysis system is embodied in the form of computer-executable instructions encoded on one or more non-transitory computer-readable media. The analysis system relies on and uses continuous data regarding one or more physiological parameters, including, but not limited to, heart rate. The continuous data used by the analysis system may be obtained or derived from an exemplary physiological measurement system disclosed herein or from a derived data source or system, such as a physiological data database. In some embodiments, the analysis system calculates, stores, and displays one or more metrics or scores related to the user's physique, health, and physical training. The scores may include, but are not limited to, an intensity score and a recovery score. The scores may be updated continuously in real time or at specified time intervals. For example, a recovery score may be determined each morning upon waking, while an intensity score may be determined in real time, after a training routine, or for the entire day.

[0193] In certain exemplary embodiments, the fitness score may be determined automatically based on physiological data of two or more users of the exemplary wearable system.

[0194] The intensity score or indicator provides an accurate measure of the cardiovascular intensity experienced by a user during a portion of a day, an entire day, or any desired period (e.g., a week or a month). The intensity score is generated and adjusted to the user's unique physiological characteristics, and may take into account, for example, the user's age, gender, anaerobic threshold, resting heart rate, maximum heart rate, etc. When determined for an exercise routine, the intensity score provides an indicator of the cardiovascular intensity experienced by the user continuously throughout the routine. When determined for a period including an exercise routine and subsequent periods, the intensity score provides an indicator of not only the cardiovascular intensity experienced by the user during the routine, but also of activities the user engaged in after the routine (e.g., couch resting, an active day shopping), which may affect recovery or exercise readiness.

[0195] In an exemplary embodiment, the intensity score is calculated based on the user's heart rate reserve (HRR) detected continuously over a desired period of time (e.g., an entire day). In one embodiment, the intensity score is an integral sum of weighted HRR detected continuously over the desired period of time. Figure 7 is a flow diagram illustrating an exemplary method for determining the intensity score.

[0196] In step 702, the continuous heart rate measurements are converted into HRR values. The time series heart rate data used in step 702 is expressed as follows:

[0197]

number

[0198] The time series of HRR measurements v(t) is defined by the following equation: where MHR is the maximum heart rate and RHR is the user's resting heart rate.

[0199]

number

[0200] In step 704, the HRR values ​​are weighted according to an appropriate weighting scheme. Cardiovascular strength, as indicated by the strength score, is defined by the following equation: where w is a weighting function for the HRR measurements.

[0201]

number

[0202] In step 706, the weighted time series of HRR values ​​are summed and normalized.

[0203]

number

[0204] That is, the weighted sum is normalized to the unit interval, i.e., [0,1], as follows:

[0205]

number

[0206] In step 708, the summed and normalized values ​​are scaled to generate a user-friendly intensity score value. That is, the unit interval is converted to have any desired distribution (e.g., arctangent, sigmoid, sinusoid, etc.) within a numerical range (e.g., a range containing 21 points from 0 to 21). In certain distributions, intensity values ​​may increase at a linear rate along the numerical range. In other distributions, intensity values ​​may increase more rapidly at the highest regions, indicating a more difficult climb up the numerical range toward the ends of the numerical range. In some embodiments, the raw intensity scores are scaled by fitting a curve to a selected group of "standard" exercise routines predefined to have specific intensity scores.

[0207] In one embodiment, a monotonic transformation of unit intervals may be performed to convert the raw HRR values ​​into user-friendly intensity scores. An exemplary scaling method is denoted as f:[0,1]→[0,1] and is performed using the following function:

[0208]

number

[0209] To generate an intensity score, the resulting value may be multiplied by a number based on the desired numerical range of the intensity score. For example, if the intensity score ranges from 0 to 21, the resulting value may be multiplied by 21.

[0210] At step 710, the intensity score value is stored in a non-transitory storage medium for retrieval, display, and use. At step 712, in some embodiments, the intensity score value is displayed in a user interface displayed on a visual display device. The intensity score value may be displayed as a numerical value and / or with the aid of a graphical tool, such as a graphical display of a numerical range of intensity scores including the current score, etc. In some embodiments, the intensity score may be presented audibly. In some embodiments, at step 712, the intensity score value may be displayed along with one or more quantitative or qualitative information about the user. This information may include, but is not limited to, whether the user exceeded their anaerobic threshold, the heart rate zones experienced by the user during the exercise routine, the difficulty of the exercise routine in the context of the user's training, the user's perceived exertion during the exercise routine, whether the user's exercise plan needs to be automatically adjusted (e.g., make it easier if the intensity score is consistently high), whether the user is likely to be sore the next day and the expected level of soreness, characteristics of the exercise routine (e.g., how difficult it was for the user, whether the exercise was intermittent or vigorous, whether the exercise was tapered, etc.), etc. In one embodiment, the analysis system may automatically generate, store, and display an exercise plan tailored to the user's intensity score.

[0211] Step 706 may use any of several exemplary static or dynamic weighting schemes that can tailor and adapt the intensity score to the user's unique physiological characteristics. In one exemplary static weighting scheme, weights applied to HRR values ​​are determined based on a static model of physiological processes. The human body uses various energy sources with varying efficiencies and advantages depending on HRR level. For example, when the anaerobic threshold (AT) is reached, the body switches to anaerobic respiration, and cells produce two adenosine triphosphate (ATP) molecules per glucose molecule, compared to 36 molecules at low HRR levels. At even higher HRR levels, there is a next threshold (CPT), at which creatine triphosphate (CTP) is used for respiration, but at an even lower efficiency.

[0212] To account for differences in the level and efficiency of cardiovascular exertion at different HRR levels, in one embodiment, the possible values ​​of HRR are divided into multiple (e.g., three) categories, sections, or levels, depending on the efficiency of cellular respiration in each category. The HRR parameter range may be divided in any suitable manner, such as piecewise, piecewise linear, or piecewise exponential. An exemplary piecewise linear division of the HRR parameter range allows each category to be weighted with strictly increasing values. This approach accurately captures the cardiovascular intensity experienced by the user. Because higher HRR values ​​make it more difficult to sustain long periods of time, the weighting function should increase for categories with increasing weights.

[0213] In one non-limiting example, the HRR parameter range may be considered to range from zero (0) to one (1) and divided into categories with strictly increasing weights. In one example, the HRR parameter range may be divided into the following categories: a first category for HRR values ​​of zero (this category is assigned a weight of zero); a second category for HRR values ​​in the range between zero (0) and the user's anaerobic threshold (AT) (this category is assigned a weight of one (1)); a third category for HRR values ​​in the range between the user's anaerobic threshold (AT) and the threshold at which the user's body uses creatine triphosphate for breathing (CPT) (this category is assigned a weight of 18); and a fourth category for HRR values ​​in the range between the creatine triphosphate threshold (CPT) and one (1) (this category is assigned a weight of 42). However, other numbers of HRR categories and different weight values ​​may be possible. Thus, in this example, the weights are defined as follows:

[0214]

number

[0215] In another exemplary embodiment of the weighting scheme, the HRR time series is iteratively weighted based on the intensity scores previously determined (e.g., the accumulated intensity scores) and the path the HRR values ​​took to reach the current intensity score. This path may be automatically detected based on past HRR values ​​and can indicate, for example, whether the user is performing high-intensity interval training (during which the intensity score rises and falls rapidly) or whether the user is taking long breaks between bouts of intermittent exercise (during which the intensity score rises after a long period of time). This path may be used to dynamically determine and adjust the weights applied to the HRR values. For example, for high-intensity interval training, the weights applied may be higher than for more traditional exercise routines.

[0216] Another exemplary embodiment of a weighting scheme uses a predictive approach by modeling weights or coefficients as coefficient estimates in a logistic regression model. In this scheme, a training data set is obtained by continuously detecting heart rate time series and other personal parameters for a group of people. The training data set is used to train a machine learning system that predicts a person's perceived cardiovascular intensity based on their heart rate and other personal data. The trained system constructs a regression model whose coefficient estimates correspond to the weights or coefficients in the weighting scheme. During the training phase, user input regarding perceived exertion is compared to intensity scores. The learning algorithm also modifies the weights based on the user's improvement or decline in fitness and qualitative feedback. This results in a unique algorithm that incorporates physiological, qualitative feedback, and quantitative data. To determine the weighting scheme for a particular user, the trained machine learning system is executed by executing computer-executable instructions encoded on one or more non-transitory computer-readable media to generate coefficient estimates. The coefficient estimates are used to weight the user's HRR time series.

[0217] Those skilled in the art will recognize that any two or more aspects of the disclosed weighting schemes may be applied separately or in combination in an exemplary method for determining an intensity score.

[0218] In one embodiment, heart rate zones quantify training intensity by weighting different levels of cardiac activity and comparing them as a percentage of maximum heart rate. Analyzing the amount of time an individual spends training at a specific percentage of their MHR may reveal the state of physical activity during training. This intensity derived from heart rate zone analysis, movement, and activity may inform the need for rest and recovery after training, for example, to minimize delayed onset muscle soreness (DOMS) and prepare for subsequent activity. As described above, MHR, heart rate zones, time above the anaerobic threshold, and HRV (heart rate variability) in the RSA (respiratory sinus arrhythmia) region, as well as personal information (such as gender, age, height, and weight) may be used in data processing.

[0219] A recovery score or index can accurately indicate a user's level of physical and health recovery after a period of physical exercise. The human autonomic nervous system controls involuntary aspects of the body's physiological functions and is generally subdivided into two branches: parasympathetic (inactivation) and sympathetic (activation). Heart rate variability (HRV), or the variation in the time intervals between heartbeats, is generally studied as the result of the interaction between these two opposing branches. Parasympathetic activation reflects input from the viscera and causes a decrease in heart rate. Sympathetic activation increases in response to stress, exercise, and illness, causing an increase in heart rate. For example, during high-intensity exercise, the sympathetic response to exercise persists long after the exercise has ended. If adequate recovery does not occur after high-intensity exercise, this imbalance typically persists until the next morning, resulting in lower morning HRV. This result should be taken as a warning sign because it indicates that the parasympathetic nervous system was suppressed overnight. While the parasympathetic nervous system is suppressed, the normal repair and maintenance processes that normally occur during sleep may also be suppressed, leaving you less prepared for the next day and making subsequent exercise more difficult.

[0220] The recovery score is developed and tailored to the user's unique physiological characteristics, taking into account, for example, the user's heart rate variability (HRV), resting heart rate, sleep quality, and recent physiological strain (indicated, for example, by the user's intensity score). In one embodiment, the recovery score is a weighted combination of the user's heart rate variability (HRV), resting heart rate, sleep quality as indicated by the sleep score, and recent strain (indicated, for example, by the user's intensity score). In one example, combining the sleep score with a performance readiness indicator (e.g., morning heart rate and morning heart rate variability) can provide the user with a complete picture of their recovery. By considering sleep and HRV, alone or in combination, the user can understand how prepared they are for exercise each day and how they arrived at their daily exercise readiness score. For example, they can understand whether a low exercise readiness score may be a predictor of poor recovery habits or an inappropriate training schedule. This insight can help the user adjust their daily activity, exercise plan, and sleep schedule to maximize their training results.

[0221] In some cases, the recovery score may take into account the subjective psychological strain felt by the user. In some cases, the subjective psychological strain may be detected from user input, for example, via a questionnaire on a mobile device or web application. In some cases, the psychological strain may be determined automatically by detecting changes in sympathetic activation based on one or more parameters, including, but not limited to, heart rate variability, heart rate, galvanic skin response, etc.

[0222] With respect to a user's heart rate variability (HRV) used to determine the recovery score, suitable techniques for analyzing HRV include, but are not limited to, time-domain methods, frequency-domain methods, geometric methods, nonlinear methods, etc. In one embodiment, the HRV metric of the root mean square difference (RMSSD) of the R-R intervals is used. The analysis system may consider the magnitude of the difference between a 7-day rolling average and a 3-day rolling average of these measurements on a particular day. Other embodiments may use Poincaré plot analysis or other suitable metrics of HRV.

[0223] The recovery score algorithm may take into account RHR along with past intensity and recovery score history.

[0224] For the user's resting heart rate, the moving average of the resting heart rate is analyzed to determine significant deviations. Taking the moving average into account is important because even healthy individuals experience significant day-to-day physiological variation. Therefore, the analysis system may perform a smoothing process to distinguish the variations from normal fluctuations.

[0225] Sleep is an inactive, yet highly restorative state in which the majority of physiological recovery processes occur. Nevertheless, small but significant amounts of recovery can occur throughout the day due to hydration, macronutrient replenishment, lactic acid removal, glycogen resynthesis, growth hormone secretion, and limited musculoskeletal repair. In assessing a user's sleep quality, the analysis system uses continuous data collected by an exemplary physiological measurement system regarding the user's heart rate, skin conductivity, ambient temperature, and accelerometer / gyroscope data during sleep to generate a sleep score. The collection and use of these four data streams enables an understanding of sleep previously only obtainable through invasive and disruptive overnight laboratory testing. For example, if skin conductivity increases in the absence of elevated ambient temperature, the wearer's heart rate is low, and there is little accelerometer / gyroscope movement, this may indicate that the wearer has fallen asleep. The sleep score is an indicator and measure of sleep efficiency (the quality of the user's sleep) and sleep duration (whether the user slept adequately). Each of these measures is determined by a combination of physiological parameters, personal habits, and daily stress / strain (intensity) inputs. Actual data measuring the time spent in various stages of sleep, combined with a longitudinal data set describing the wearer's recent daily history and personal habits, can assess the quality of sleep achieved by the user. The sleep score is designed to model sleep quality in terms of sleep duration and history. Therefore, the continuous monitoring properties of the exemplary physiological measurement system disclosed herein can be leveraged by considering each sleep period in terms of biologically determined sleep needs, pattern-determined sleep needs, and historically determined sleep debt.

[0226] The recovery score value and sleep score value are stored in a non-transitory storage medium for retrieval, display, and use. In some embodiments, the recovery score value and / or sleep score value may be displayed in a user interface displayed on a visual display device. The recovery score value and / or sleep score value may be displayed numerically and / or with the aid of a graphical tool, such as a graphical display of a numerical range of recovery scores including the current score. In some embodiments, the recovery score value and / or sleep score may be presented audibly. In some embodiments, the recovery score value may be displayed along with one or more quantitative or qualitative information about the user. This information may include, but is not limited to, whether the user is sufficiently recovered, the level of activity the user is ready to perform, whether the user is ready to perform an exercise routine of a particular desired intensity, whether the user should rest and the recommended duration of the rest, whether the user's exercise plan needs to be automatically adjusted (e.g., to simplify if the recovery score is low), etc. In one embodiment, the analysis system may automatically generate, store, and display a customized exercise plan based on the user's recovery score alone or in combination with an intensity score.

[0227] As described above, sleep performance metrics may be determined based on parameters such as the number of hours of sleep, sleep onset latency, and number of sleep disturbances. In this manner, a score can compare the duration and quality of a tactical athlete's sleep to the tactical athlete's changing sleep needs (e.g., hours based on recent strain, habitual sleep needs, illness symptoms, and sleep debt). For example, soldiers' sleep needs may change dynamically, and it may be important to consider total sleep hours in relation to the amount of sleep they may have needed. According to one aspect, providing accurate sleep and sleep performance sensors allows for assessment of a particular user's sleep throughout their day and in the context of their lifestyle.

[0228] FIG. 8 is a flow diagram illustrating an exemplary method for a user to use intensity and recovery scores. In step 802, the wearable physiological measurement system begins determining a heart rate variability (HRV) measurement based on continuous heart rate data collected by the exemplary physiological measurement system. In some cases, collecting heart rate data for several days may be necessary to obtain an accurate baseline for HRV. In step 804, the analysis system can generate and display an intensity score for the entire day or exercise routine. In some cases, the analysis system may display quantitative and / or qualitative information corresponding to the intensity score. FIG. 9 shows an exemplary display of an intensity score index represented by a circular graphical component, indicating an exemplary current score of 19.0. The graphical component may indicate the difficulty of the exercise corresponding to the current score, e.g., selected from full intensity, nearly full intensity, very hard, hard, moderate, light, active, light activity, no activity, sleeping, etc. The display may indicate, for example, that the intensity score corresponds to a good tapering exercise routine, that the user did not exceed the anaerobic threshold, and that the user will experience little or no muscle soreness the next day.

[0229] In an exemplary embodiment, in step 806, the analysis system may automatically generate or adjust an exercise routine or exercise plan based on the user's actual intensity score or desired intensity score. For example, based on input of the user's actual intensity score, desired intensity score (which is higher than the actual intensity score), and a first exercise routine (e.g., walking) that the user is currently performing, the analysis system may recommend a second, different exercise routine (e.g., running) that is typically associated with a higher intensity score than the first exercise routine.

[0230] At step 808, at any given time during the day (e.g., every morning), the analysis system may generate and display a recovery score. In some cases, the analysis system may display quantitative and / or qualitative information corresponding to the intensity score. In an exemplary embodiment, for example, at step 810, the analysis system may determine whether the recovery score is greater than (or equal to) a first predetermined threshold (e.g., in some examples, about 60% to about 80%) indicating that the user is recovered and ready to exercise. If greater than the first predetermined threshold, at step 812, the analysis system may indicate that the user is ready to perform an exercise routine at a desired intensity, or that the user is ready to perform an exercise routine that is more difficult than the previous day's routine. Otherwise, at step 814, the analysis system may determine whether the recovery score is less than (or equal to) a second predetermined threshold (e.g., in some examples, about 10% to 40%) indicating that the user is not recovered. If less than the second predetermined threshold, at step 816, the analysis system may indicate that the user should not exercise and should take an extended rest. In some cases, the analysis system may extend the recommended rest period. Otherwise, in step 818, the analysis system may indicate that the user may exercise according to their exercise plan, but with caution. In some cases, the threshold may be adjusted based on the desired intensity at which the user wishes to exercise. For example, the higher the intensity score on the plan, the higher the threshold may be.

[0231] Figure 10 shows an exemplary display of a recovery score index shown in a circular graphic component. A first threshold is shown as 66% and a second threshold is shown as 33%. Figures 11A-11C show the recovery score graphic component with exemplary recovery scores and corresponding qualitative information for the recovery scores.

[0232] Optionally, in an exemplary embodiment, the analysis system may automatically generate or adjust an exercise routine or plan based on the user's actual recovery score (e.g., recommending lighter exercise on days when the user is not sufficiently recovered). A combination of intensity and recovery scores may be used in this process.

[0233] In some embodiments, the analysis system may determine and display intensity and / or recovery scores for multiple users in a comparative manner, allowing users to compete against other users in their exercise routines based on comparative intensity scores.

[0234] III. Exemplary Displays and User Interfaces

[0235] Exemplary embodiments also provide a vibrant, interactive online community for viewing and sharing physiological data among users. Exemplary systems have the ability to wirelessly stream physiological information to an online website, either directly or via a mobile device application. The website may allow users to monitor their fitness results, share information with teammates and coaches, compete against other users, and earn status. Both the wearable system and the website allow users to receive feedback about their day, enabling recovery and performance evaluation. One aspect is directed to providing an online website for health and fitness monitoring. In some embodiments, the website may be a social networking site. The website may allow users, such as young athletes, to monitor their fitness results, share information with teammates and coaches, compete against other users, and win prizes. Users include individuals receiving health and fitness monitoring, and may include, for example, individuals wearing the bracelets disclosed herein, athletes, sports team members, personal trainers, coaches, etc. In some embodiments, users may select their trainer from a list to comment on their performance.

[0236] In some embodiments, the website may be configured to provide an interactive user interface. The website may be configured to display results based on an analysis of the physiological data received from one or more devices. The website may be configured to provide a competitive way for users to compare themselves to one another, ultimately providing a more interactive experience for users. For example, in some embodiments, rather than simply comparing a user's physiological data and performance to that user's past performance, the website may allow users to compete against other users and compare their performance to that of other users.

[0237] In some embodiments, the website may be a mobile website or a mobile application. In some embodiments, the website may be configured to communicate data to other websites or applications.

[0238] An exemplary website may include a simple, free sign-up process during which a user can create an account using their name, account name, email address, home address, height, weight, age, and a unique code assigned to their wearable physiological measurement system. The unique code may be assigned, for example, to the wearable system itself or to a packaged kit. Once registration is complete, continuous physiological data received from the user's system may be continuously retrieved in real time and automatically displayed on a web page associated with the user. Additionally, users may add information to their profile, such as a picture, favorite activities, and sports teams, and users may search for and share information with teammates and friends on the website.

[0239] 12A-14B illustrate an exemplary user interface 1200 for displaying user-specific physiological data displayed on a visual display device. In some embodiments, the user interface 1200 can take the form of a web page. Those skilled in the art will appreciate that the information in FIGS. 12A-14B is a non-limiting example. The user interface 1200 may include a summary panel 1202 that includes the user's identifying information 1204 (e.g., real name or account name), and optionally, the summary panel 1202 may have an image or photo corresponding to the user. The summary panel 1202 also displays the user's current intensity score 1206 and current recovery score 1208. In some embodiments, the summary panel 1202 may display the number of calories 1210 the user has burned that day and the number of hours 1212 the user has slept the previous night.

[0240] User interface 1200 may further include various panels for displaying information about a user's workout, such as a workout panel 1214 accessible using tab 1216, a day panel 1318 accessible using tab 1220, and a sleep panel 1422 accessible using tab 1224. The workout panel, day panel, and sleep panel may have the same or different feedback panels associated with them. These panels may allow a user to select and customize one or more information panels that are displayed on their user interface display.

[0241] The workout panel 1214 may display quantitative information about the user's fitness and exercise routine, such as a graph 1230 of the user's continuous heart rate during exercise, statistics 1232 about maximum heart rate, average heart rate, exercise time, steps, and calories burned, the zone 1234 in which maximum heart rate decreased during exercise, and a graph 1236 of intensity score over a period of time (e.g., seven days).

[0242] A feedback panel 1238 associated with the workout panel 1214 may display information about the intensity score during a selected time period and the exercise routine the user performed during that time period. The information may include, but is not limited to, quantitative information, qualitative information, feedback, and recommendations for future exercise routines. Along with the intensity score, the feedback panel 1238 may display a qualitative summary 1240 of the score. The summary may indicate, for example, whether the user exercised above their anaerobic threshold for a significant period of time, or whether the exercise may cause muscle soreness or stiffness. The feedback panel 1238 may display one or more tips 1242 for adjusting the exercise routine based on an analysis of quantitative health parameters monitored during the exercise routine. For example, a tip may indicate that the exercise routine started too quickly and the user needs to warm up longer. In some cases, when a tips sub-panel 1242 is selected, a corresponding indicator 1244 may be displayed on the heart rate graph 1230.

[0243] The feedback panel 1238 may also display qualitative information 1245 about the user's exercise routine based on an analysis of quantitative health parameters monitored during the exercise routine, such as a comparison of that day's exercise routine with the user's past exercise data. Such information may indicate, for example, that the user's maximum heart rate for that day's exercise was the highest ever recorded, that the user's steps for that day were the lowest ever recorded, that the user burned a lot of calories, and that more calories could be burned if the exercise intensity was reduced. The feedback panel 1238 may also display warning indicators 1246 to alert the user to potential future health events, such as the possibility of muscle soreness (e.g., if the intensity score is higher than a predetermined threshold).

[0244] An example analysis system can analyze the information displayed in the workout panel 1214 to determine whether a user performed a particular exercise routine or activity. As an example, if the step count is low and the calorie burn and heart rate are high, the system may determine that the user likely rode a bike that day. In some cases, the feedback panel 1238 may prompt the user in the user field 1248 to confirm whether the user actually performed the activity. This user input may be displayed and / or used to further understand the user's fitness and exercise routine.

[0245] The day panel 1318 may include information about the user's health parameters for that day, including, but not limited to, the number of calories burned and consumed 1350 (which may be based on user input about foods eaten), a graph of the continuous heart rate for that day 1354, statistics about the user's resting heart rate and steps for that day 1356, a graph of calories burned for that day and other days 1358, etc.

[0246] In some cases, the analysis system may analyze physiological data (e.g., heart rate data) to estimate the amount of sleep, activity, and exercise time for that day. A feedback panel 1362 associated with the day panel 1318 may display these times 1364. In some cases, the feedback panel 1362 may display the net calories 1366 burned by the user for that day. The feedback panel 1362 may also display qualitative information 1368 about the user's exercise routine based on an analysis of quantitative health parameters monitored during the exercise routine. This information may indicate, for example, that the user was stressed at a particular point in the day (e.g., sweating a lot despite low activity), that the user's maximum heart rate for that day's exercise was the highest ever recorded, that the user's steps for that day were the lowest ever recorded, that the user burned a lot of calories, and / or could burn more calories by reducing the intensity of their exercise. The feedback panel 1362 may also display warning indicators 1370 to alert the user to upcoming health events, such as tachycardia, susceptibility to illness, overtraining (e.g., elevated resting heart rate for several days), etc.

[0247] An example analysis system can analyze the information displayed in the day panel 1318 to determine whether a user performed a particular exercise routine or activity. As an example, if the heart rate is elevated despite little activity, the system may determine that the user may have had coffee at that time. In some cases, the feedback panel 1362 may prompt the user in the user field 1372 to confirm whether the user actually performed the activity. This user input may be displayed and / or used to further understand the user's health and exercise routine.

[0248] The sleep panel 1422 may include information about the user's health parameters while sleeping, such as, but not limited to, a graph 1473 overlaying heart rate and movement during sleep, statistics 1474 regarding maximum heart rate, minimum heart rate, the number of times the user woke up during sleep, and average movement during sleep, a sleep cycle indicator 1476 showing time spent awake, time in light sleep, time in deep sleep, and time in REM sleep, and a sleep duration graph 1478 showing the number of hours slept over a period of time.

[0249] A feedback panel 1480 associated with the sleep panel 1422 may display information about the user's sleep, including, but not limited to, quantitative information, qualitative information, feedback, and recommendations for future exercise routines. The feedback panel 1480 may display a sleep score and / or number of hours of sleep, along with a qualitative summary 1482 of the score. The summary may include, for example, whether the user slept enough, whether the sleep was efficient or inefficient, whether and how much the user moved around during sleep, etc. The feedback panel 1480 may display one or more tips 1484 for adjusting sleep based on an analysis of the quantitative health parameters monitored during sleep, such as indicating that the user woke up multiple times during sleep or suggesting that the user try sleeping on their side instead of their back.

[0250] The feedback panel 1480 may further display qualitative information 1486 about the user's sleep based on an analysis of the quantitative health parameters monitored during the exercise routine. Such information may indicate, for example, that the user's maximum heart rate for that day's exercise was the highest recorded while sleeping. The feedback panel 1480 may also display warning indicators 1488 to alert the user to potential future health events, such as signs of overtraining or recommendations to get more sleep (e.g., if the user wakes up multiple times during sleep or moves around during sleep).

[0251] User interface 1200 may include user input fields 1290 to allow the user to display their feelings (e.g., activity performed, perceived exertion, energy level, performance, etc.). User interface 1200 may further include user input fields 1292 to allow the user to display other facts about their exercise routine, such as comments about what the user was doing at a particular point in the exercise routine, along with a link 1294 to the corresponding point on heart rate graph 1230. In some embodiments, the user may specify the route and / or map location where the exercise routine was performed.

[0252] In an exemplary embodiment, a user may also compare their quantitative and / or qualitative physiological data with that of one or more other users. The user may be presented with a user selection component representing other users with viewable data. Hovering the pointer over the user selection component (e.g., an icon representing the user) displays a snapshot of the user's information in a pop-up component, and clicking the user selection component opens a full user interface displaying the user's information. In some cases, the user selection component may include specific data for a particular user (e.g., a graphical element indicating the user's strength score) surrounding an image representing the user. The user selection component may be presented in a grid format, as shown, or in a linear list for easier sorting. The users displayed in the user selection component may be sorted and / or ranked based on any criteria, such as strength score or which users experience muscle soreness. Users may leave comments on other users' pages.

[0253] Similarly, users can select privacy settings to specify which portions of their data are visible to other users. Because the wearable systems described herein support true continuous monitoring, users may want to carefully control whether and when data is transmitted wirelessly, whether and when data is stored in remote data repositories, and whether and when data is shared with other users. The privacy switches described herein can be effectively used to toggle between various privacy settings or to explicitly select private or restricted times when monitoring should not occur.

[0254] FIG. 15 is a flow diagram illustrating a method for selecting a mode for acquiring heart rate data.

[0255] As shown in step 1502, method 1500 may include providing a strap with a sensor and a heart rate monitoring system. The strap may be shaped and sized to fit a portion of the body. For example, the strap may be any of the straps described herein, including, but not limited to, a bracelet. The heart rate monitoring system may be configured with two or more different modes for detecting the heart rate of a wearer of the strap. The modes may include the use of optical detectors (e.g., photodetectors), light emitters, motion sensors, processing modules, algorithms, other sensors, peak detection techniques, frequency domain techniques, variable optical properties, non-optical techniques, and the like.

[0256] As shown in step 1504, method 1500 may include detecting a signal from a sensor. The signal may be detected by one or more sensors, which may include any of the sensors described herein. The signal may include, but is not limited to, one or more signals related to the user's heart rate, other physiological signals, optical signals, motion-based signals, signals based on environmental factors, status signals (e.g., battery life), and historical information, etc.

[0257] As shown in step 1506, method 1500 may include determining a state of the heart rate monitoring system. The state may be determined based on the signal. The state may include, but is not limited to, heart rate detection accuracy determined using statistical analysis to ensure confidence in accuracy, power consumption, battery charge level, user activity, sensor location or sensor movement, environmental or situational conditions (e.g., ambient light conditions), physiological state, active state, inactive state, etc. This may include detecting a change in state, responsively selecting another mode from among two or more different modes, and storing additional continuous heart rate data obtained using at least one of the two or more different modes.

[0258] As shown in step 1508, method 1500 may include selecting one of two or more different modes for detecting a heart rate based on a condition. For example, the method may automatically and selectively activate one or more light emitters based on a user's motion state to identify the user's heart rate. The system may also or alternatively determine the type of sensor to use at a given time based on the level of motion, skin temperature, heart rate, etc. The system may also select the type of sensor to use when monitoring the user's heart rate based on a combination of these factors. A processor or the like may be configured to select one of the modes. For example, if the condition is the accuracy of heart rate detection, which is determined using statistical analysis to ensure confidence in the accuracy, the processor may be configured to select another mode from the modes when the confidence falls below a predetermined threshold.

[0259] As shown in step 1510, method 1500 may include storing the continuous heart rate data using one of two or more different modes. This may include communicating the continuous heart rate data from the strap to a remote data repository. This may additionally or alternatively include storing the data locally, e.g., in memory included with the strap. The memory may be removable (e.g., via a data card, etc.) or may be permanently attached / integrated to the strap or a component thereof. The storage of the data (e.g., heart rate data) may be for the user's personal use, e.g., in a private environment; or, in a shared environment (e.g., on a social networking site, etc.), the data may be shared. Method 1500 may further include the use of a user-operable privacy switch to controllably limit communication of some of the data (e.g., to a remote data repository).

[0260] FIG. 16 is a flow diagram of a method for assessing recovery and making exercise recommendations.

[0261] As shown in step 1602, method 1600 may include monitoring data from a wearable system. The wearable system may be a continuous monitoring physiological measurement system worn by a user. The data may include heart rate data, other physiological data, summary data, motion data, fitness data, activity data, or any other data described herein or envisioned by one of ordinary skill in the art.

[0262] As shown in step 1604, method 1600 may include detecting athletic activity. This may include automatically detecting the user's athletic activity. The athletic activity may be detected using one or more sensors as described herein. The athletic activity may be transmitted to a server, such as a server that performs step 1606 described below.

[0263] As shown in step 1606, method 1600 may include generating an assessment of the athletic activity, which may include generating a quantitative assessment of the athletic activity. Generating the quantitative assessment of the athletic activity may include analyzing the athletic activity on a remote server. Generating the quantitative assessment may include using an algorithm described herein. Method 1600 may further include generating periodic updates to the user regarding the athletic activity. Method 1600 may further include determining a qualitative assessment of the athletic activity and communicating the qualitative assessment to the user.

[0264] As shown in step 1608, method 1600 may include detecting a recovery state. This may include automatically detecting the user's physical recovery state. The recovery state may be detected using one or more sensors as described herein. The recovery state may be transmitted to a server, such as a server that performs step 1610 described below.

[0265] As shown in step 1610, method 1600 may include generating a recovery status assessment. This may include generating a quantitative assessment of the physical recovery status. Generating the quantitative assessment may include using an algorithm described herein. Generating the quantitative assessment of the physical recovery status may include analyzing the physical recovery status on a remote server. Method 1600 may further include generating periodic updates to the user regarding the physical recovery status. Method 1600 may further include determining a qualitative assessment of the recovery status and communicating the qualitative assessment to the user.

[0266] As shown in step 1612, method 1600 may include analyzing the assessment, i.e., analyzing a quantitative assessment of athletic activity and a quantitative assessment of physical recovery. The analysis may include use of one or more of the algorithms described herein, statistical analysis, etc. The analysis may include use of a remote server.

[0267] As shown in step 1614, method 1600 may include generating recommendations. This may include automatically generating recommendations regarding modifications to the user's exercise routine based on the analysis performed in step 1612. Additionally or alternatively, this may include determining a qualitative assessment of exercise activity and / or recovery status and communicating the qualitative assessment to the user. The recommendations may be generated on a remote server. The recommendations may be communicated to the user via email, presented to the user through a web page or other communication interface, etc. The generation of recommendations may be based on the number of exercise and rest cycles.

[0268] The above method 1600, or any of the methods described herein, may additionally or alternatively be implemented on a computer program product including non-transitory computer-executable code embodied in a non-transitory computer-readable medium and executed on one or more computing devices to perform the steps of the method. For example, code may be provided to perform various steps of the methods described herein.

[0269] 17 is a flow diagram illustrating a method for detecting heart rate variability during sleep. Method 1700 can be used in conjunction with any of the devices, systems, and methods described herein, for example, by operating a wearable continuous physiological monitoring device to perform the following steps: The wearable continuous physiological monitoring system may include, for example, a processor, one or more light-emitting diodes, one or more photodetectors configured to acquire heart rate data from a user, and one or more other sensors to assist in detecting sleep stages. Generally, method 1700 is intended to measure heart rate variability during the final stages of sleep before waking in order to obtain a consistent and accurate baseline for calculating a physical regeneration score.

[0270] As shown in step 1702, method 1700 may include detecting a sleep state of the user, which may include any form of continuous or periodic monitoring of the sleep state, for example, using any of the various sensors and algorithms generally described herein.

[0271] Sleep states (also referred to as "sleep phases," "sleep cycles," "sleep stages," etc.) may include rapid eye movement (REM) sleep, non-rapid eye movement (NREM) sleep, or any of the states / stages within them. Sleep states may include various stages of non-rapid eye movement (NREM) sleep, from stages 1 through 3. Stage 1 of non-rapid eye movement (NREM) sleep generally involves a state in which the eyes are closed but the individual can be easily awakened. Stage 2 of non-rapid eye movement (NREM) sleep generally involves a state in which the individual is in a lighter sleep state, i.e., a state in which the individual's heart rate slows and body temperature drops in preparation for deeper sleep. Stage 3 of non-rapid eye movement (NREM) sleep generally involves a state in which the individual cannot be easily awakened. Stage 3 is often referred to as delta sleep, deep sleep, or slow-wave (i.e., the high-amplitude, low-frequency brain waves typically observed during this stage) sleep. Slow-wave sleep is considered the most restful form of sleep, reducing subjective sleepiness and restoring the body.

[0272] REM sleep, on the other hand, typically occurs one to two hours after falling asleep. REM sleep may include various periods, stages, or phases, all of which may be included in the sleep states detected as described herein. During REM sleep, breathing becomes rapid, irregular, and shallow, and the eyes may twitch rapidly (hence the term "rapid eye movement," or "REM"). Muscles in the limbs may also become temporarily paralyzed. During this stage, brain waves increase to levels typically experienced while awake. Heart rate, cardiac pressure, cardiac output, and arterial blood pressure may also become irregular as the body enters REM sleep. This is the sleep state in which most dreams occur, and if awakened during REM sleep, individuals can usually recall their dreams. Most people experience three to five REM sleep intervals each night.

[0273] Homeostasis is the balance between sleep and wakefulness, and maintaining proper homeostasis can be beneficial to human health. Sleep deprivation, commonly referred to as sleep insufficiency, tends to cause slowed brain waves, decreased attention, increased anxiety, memory impairment, mood disorders, and overall mental, emotional, and physical fatigue. Sleep debt (the effects of not getting enough sleep) can lead to a decline in the ability to perform higher cognitive functions. A person's circadian rhythm (i.e., a biological process that exhibits endogenous, synchronized oscillations over approximately 24 hours) may contribute to ensuring an optimal amount of sleep. Therefore, sleep in general may be usefully monitored as a surrogate for physical recovery. However, a person's heart rate variability at specific moments during sleep (i.e., the final stage of sleep before a wake event) can further provide an accurate and consistent basis for objectively calculating a recovery score after a sleep period.

[0274] As described above, monitoring a user's sleep can detect various sleep states, transitions, and other sleep-related information. For example, a device can monitor / detect the duration of sleep states, transitions between sleep states, the number of sleep cycles or specific states, the number of transitions, the number of wake-up events, transitions to wake-up states, etc. Sleep states can be monitored and detected using various strategies and sensor configurations depending on the underlying physiological phenomenon. For example, body temperature can be advantageously correlated with various sleep states and transitions. Similarly, galvanic skin response can be correlated with sweat activity and various sleep states. Monitoring any of these (e.g., with a galvanic skin response sensor) can determine sleep state. Body movement can also be easily monitored using an accelerometer or the like and can be used to detect wake-ups or other activities involving body movement. In another aspect, heart rate activity itself, alone or in combination with other sensor data, may be used to infer various sleep states and transitions. Other sensors, such as an electroencephalogram (EEG) monitor, a pupil monitor, etc., can also be used in addition to or instead of monitoring sleep activity. However, the ability to incorporate these types of detection into continuous wearable physiological monitoring devices may be somewhat limited depending on the envisioned configuration.

[0275] As shown in step 1704, method 1700 may include substantially continuous monitoring of the user's heart rate using a continuous physiological monitoring system. Continuous heart rate monitoring has been described in considerable detail above and will not be repeated here. However, this may include raw sensor data, heart rate data or peak data, and heart rate variability data over some historical period, which can then be correlated with various sleep states and activities.

[0276] As shown in step 1706, method 1700 may include recording the heart rate as heart rate data. This may include storing the heart rate data in raw or processed form on the device or transmitting and storing the data at a local or remote location. In one aspect, the data may be stored as peak-to-peak data or other semi-processed form without calculating heart rate variability. This may be useful as a technique to conserve processing resources in various situations, such as when only heart rate variability at a particular time point is of interest. The data may be logged in raw or semi-processed form, and then heart rate variability at the relevant time point may be calculated after the relevant time point is identified.

[0277] As shown in step 1710, method 1700 may include detecting a wake-up event upon the user's transition from a sleep state to a wake-up state. It should be understood that a wake-up event may occur as a result of a natural termination of sleep, such as after a full night's rest, or may occur in response to an external stimulus that awakens the user before the natural sleep cycle is complete. Regardless of the triggering event, the wake-up event may be detected via the various physiological changes described above or using other suitable techniques. While the focus herein is on wearable continuous monitoring devices, it will be understood that the device may also receive additional input from external devices, such as a camera (for motion detection) or an infrared camera (for body temperature detection), which may be used to help accurately assess various sleep states and transitions.

[0278] Thus, a wearable continuous physiological monitoring system may generally detect a wake-up event using one or more sensors including, for example, one or more of an accelerometer, a galvanic skin response sensor, an optical sensor, etc. For example, in one aspect, a wake-up event may be detected using a combination of motion data and heart rate data.

[0279] As shown in step 1712, method 1700 may include calculating the user's heart rate variability during the final stages of sleep prior to the wake-up event based on the heart rate data. While a history of wake-up events and sleep states is useful information for assessing regeneration, method 1700 described herein specifically contemplates the use of heart rate variability during the final stages of sleep as a consistent basis for calculating a device user's regeneration score. Thus, step 1712 may also include detecting a period of slow-wave sleep immediately prior to the wake-up event and determining the end of a slow-wave sleep or deep sleep episode immediately prior to the wake-up event.

[0280] It will be appreciated that the final stage of sleep immediately preceding a natural wake-up event may be slow-wave sleep. However, if a sleeper awakens early, the final stage may instead include the last recorded episode of REM sleep or another sleep stage immediately preceding the wake-up event. This moment, i.e., the end of the final stage of sleep before waking, is the point at which heart rate variability data provides the most accurate and consistent indication of physical recovery. Therefore, by identifying an appropriate point in time, past heart rate data (in any form) can be used in the above techniques to calculate corresponding heart rate variability. Furthermore, it should be noted that the time for this calculation can be selected with varying granularity depending on the ability to accurately detect the final stage of sleep and its end. Thus, for example, this time could be a predetermined time before waking, the end of slow-wave sleep, or a predetermined length of time before the end of slow-wave sleep is detected or estimated. In another aspect, an average heart rate variability or similar indicator may be determined for any number of discrete measurements within a window around the time of interest.

[0281] As shown in step 1714, method 1700 may include calculating the duration of sleep states. The quantity and quality of sleep may be very closely related to physical recovery, and therefore the duration of sleep states may be used to calculate a recovery score.

[0282] As shown in step 1718, method 1700 may include evaluating the quality of heart rate data for slow-wave sleep periods, such as slow-wave sleep periods occurring immediately before a wake-up event, using a data quality metric. As previously discussed, the quality of heart rate measurements can change over time for a variety of reasons. Therefore, the quality of the heart rate data may be evaluated before selecting a particular moment or window of heart rate data for calculating heart rate variability, and method 1700 may include using this quality data to select an appropriate value for calculating the resilience score. For example, method 1700 may include calculating heart rate variability for a window of predetermined duration within the slow-wave sleep period having the highest quality heart rate data according to the data quality metric.

[0283] As shown in step 1720, method 1700 may include calculating a resilience score for the user based on heart rate variability obtained from the final stage of sleep. The calculation may be based on other data sources. For example, the resilience score may be calculated based on sleep duration, information about the detected sleep stage or stages (e.g., the amount of time in a particular stage), information about a recent slow-wave sleep period or another sleep period / state, information from a GSR sensor or other sensor, etc. Method 1700 may further include calculating an additional resilience score after one or more other wake events for the user and comparing it to previously calculated resilience scores. The actual calculation of the discovery score has been described in considerable detail above and will not be repeated here. However, it should be noted that using heart rate variability measurements obtained from the final stage of sleep provides an accurate and consistent basis for assessing the user's physical resilience state after a period of sleep.

[0284] As shown in step 1730, the method 1700 may include calculating a sleep score and communicating this score to the user.

[0285] In one embodiment, a sleep score can be a measure of past sleep performance. For example, a sleep performance score may quantify the ratio of sleep time during a particular rest period to sleep time needed, using a numerical range from 0 to 100. In this numerical range, if a user slept 6 hours and needed 8 hours of sleep, the sleep performance might be calculated as 75%. A sleep performance score may start from one or more assumptions about sleep need based on age, gender, health status, fitness level, habits, genes, etc., and may be adjusted to the actual sleep patterns measured for an individual over time.

[0286] The sleep score may additionally or alternatively include a sleep need score or other objective indicator that estimates the amount of sleep the user of the device will need during the next sleep period. Generally, the score may have any suitable quantitative representation, such as, for example, a numerical value within a predetermined range (e.g., 0-10, 1-100, or other suitable range) or the number of hours of sleep the user should aim for. In another aspect, the sleep score may be calculated as the number of additional hours of sleep needed in addition to the user's usual amount of sleep.

[0287] This score may be calculated using any suitable inputs, reflecting considerations such as current sleep deprivation, strain or intensity measurements over a given past interval, naps or other rest periods, etc. Various factors, such as physiological attributes such as age, sex, health status, and genes, as well as daytime activity, stress, naps, sleep deprivation, and sleep deprivation, may affect actual sleep need. The sleep deprivation itself may be calculated based on sleep need and actual sleep performance (quality, duration, wake interval, etc.) over a given past period. In one embodiment, the objective scoring function for sleep need may have a model of the form:

[0288] Sleep need = baseline + f1 (burden) + f2 (debt) - nap

[0289] In general, this calculation aims to estimate the ideal amount of sleep to achieve optimal rest and recovery during the next sleep period. The actual amount of time devoted to sleep may be somewhat longer due to factors such as the time it takes to fall asleep and short periods of awakening. This may be explicitly factored into the sleep need calculation, or it may be left alone if the user chooses to manage their sleep habits accordingly.

[0290] Generally, baseline sleep represents the typical amount of sleep a user requires on a typical rest day (e.g., a day without strenuous exercise or training). As mentioned above, this can depend on a variety of factors and can be estimated or measured in any way appropriate for a particular individual. The strain component f1 (strain) is assessed based on the intensity of physical activity the previous day and typically increases sleep needs. If intensity or strain is measured on an objective numeric range from 0 to 21, the strain calculation takes the form below, where the additional sleep required for strain i is calculated in minutes as follows:

[0291]

number

[0292] Sleep debt f2 (debt) can generally be a measure of the carryover sleep requirement not met the previous day. It may be scaled and capped depending on the individual's sleep characteristics and general information about long-term sleep deprivation and recovery. Naps can also be taken into account directly by correcting sleep requirement for naps taken, or by calculating a nap factor that is scaled or manipulated or calculated to more accurately track the actual impact of naps on future sleep needs.

[0293] Regardless of the method of calculation, the sleep need can be communicated to the user, for example, by displaying the sleep need on a wrist-worn physiological monitoring device, or by sending an email, text message, or other alert to the user and displaying it on an appropriate device.

[0294] Described herein are physiological monitoring devices, systems, and methods for detecting and analyzing periods during which a user is asleep or resting. However, sleep opportunity (i.e., time devoted to sleep) is typically achieved in only approximately 90%-95% of the total available time opportunity. The remaining 5%-10% of sleep opportunity may be lost due to short sleep interruptions lasting from a few seconds to a few minutes. This poses conflicting challenges for sleep assessment. On the one hand, subjects reviewing sleep metrics may prefer to perform quantitative analysis as soon as possible so that data is available immediately after waking. On the other hand, if the sleep interruption is temporary and the subject intends to go back to sleep, initiating computationally expensive server-side sleep analysis may be wasteful. To provide timely sleep analysis for subjects who intend to wake up while minimizing the computational burden associated with discarded data, the techniques described herein can advantageously be used to predict a subject's sleep intention. This allows for a more rapid and accurate assessment of whether a wake-up event reflects a temporary sleep interruption or the subject's intention to wake up.

[0295] As described herein, a system may be configured to detect a subject's sleep intent to determine the likelihood that a given wake-up event is associated with an intention to stay awake. This analysis may apply heuristics and / or rules derived from historical data of a user or a population, such as the user's past sleep patterns, the sleep patterns of other users, or a combination thereof. In one aspect, sleep intent may be detected using probabilistic analysis of the user's historical data, including information such as time of day, day of the week, and seasonal factors. Detection of sleep intent may also be aided by probabilistic analysis of data from a broader population, such as through similarity models or population-wide analysis.

[0296] FIG. 18 is a flow diagram illustrating a method for detecting sleep intent. Method 1800 may be used in any of the methods or systems described herein to improve sleep scoring. In general, the techniques described below aim to quickly and accurately distinguish between a user who is waking up and a user who is returning to sleep. More specifically, the techniques aim to conserve processing resources used to assess sleep quality when a user is more likely to remain awake and review sleep metrics. This is particularly advantageous, for example, in systems where sleep processing is computationally expensive or where sleep processing is performed by remote processing resources.

[0297] As shown in step 1802, method 1800 may include detecting the beginning of a sleep interval. Generally, a device, such as any of the devices described herein, may acquire data, such as heart rate data, continuously, substantially continuously, or intermittently and analyze this data to detect the beginning of a sleep interval. In this context, a sleep interval may include an interval that a user desires to analyze (e.g., sleep quality or duration) or an interval suitable for evaluation and automatic detection. For example, a sleep interval may include an entire night's sleep, e.g., the sleep time measured from the time the user gets into bed and falls asleep to the time the user gets out of bed and begins the next day's activities. This interval may also or instead include substantial intervals during which the user's inactivity or sleep was detected. In another aspect, a sleep interval may include other sleep sessions, such as naps. A sleep interval may include a user's sleep opportunity (time spent sleeping). For example, the beginning of a sleep interval may be when the user assumes a sleep position (e.g., when lying down in bed) before first falling asleep.

[0298] It will be appreciated that a sleep interval may include subsequent intermittent events, such as when a user briefly wakes up and then falls asleep again. That is, a sleep interval may include one or more subintervals, such as when a user wakes up and then falls asleep again, or when the user is in a different sleep mode. To this end, method 1800 may additionally or alternatively include detecting the beginning and end of such subintervals, particularly where such information is relevant to assessing sleep quality or quantity and / or where such information is useful for accurately identifying the beginning of a sleep interval to be analyzed and / or used in analyzing historical data.

[0299] It will be appreciated that numerous sleep detection techniques based on the physiology of sleep are known in the art. Thus, for example, changes in body temperature, sweating, movement, heart rate, respiratory rate, eye movement, blood pressure, blood oxygen level, brain waves, etc., can all be used, either alone or in combination, to detect not only the onset of sleep but also transitions between various sleep stages (e.g., light sleep, deep sleep, REM sleep, etc.). Each of these sleep stages has objectively measurable physical characteristics. Some or all of these physical characteristics can be detected by a wearable physiological monitoring device, such as any of the devices described herein, and used to detect the onset of sleep at the beginning of a sleep segment, as well as multiple sub-segments within a sleep cycle and transitions between various sleep stages.

[0300] As shown in step 1804, method 1800 may include acquiring sensor data using a physiological monitor worn by the user, for example, during the sleep segment. The physiological monitor may be a wearable physiological monitor, such as any of the devices described herein, including, but not limited to, a bracelet or other wearable strap that includes one or more sensors for acquiring physiological data. For example, the sensors may include light-emitting diodes and optical sensors (e.g., to acquire photoplethysmography data), accelerometers (to acquire motion data), thermocouples (to acquire temperature data), microphones (to acquire acoustic data), capacitance sensors, etc. (to acquire electrodermal response data), etc. It should be understood, however, that the techniques described herein are applicable more generally to any physiological monitoring system that may advantageously conserve computing resources by accurately identifying sleep intent, or more specifically, by identifying when a user intends to wake up and stay awake after a sleep segment.

[0301] As shown in step 1806, method 1800 may include detecting a wake-up event with a physiological monitor. A wake-up event may be any event that objectively and measurably indicates that a user has woken up from a sleep period. A wake-up event may include a single event, such as a change in heart rate or detected movement, or may include a composite event based on multiple measurable events that collectively indicate an awakening from a sleep state.

[0302] Numerous techniques may be used alone or in combination to detect a wake-up event. For example, detecting a wake-up event may include detecting a bodily movement inconsistent with sleep, such as sitting up, getting out of bed, lifting a limb above a predetermined height, or movement of a body part above or below a predetermined velocity or acceleration threshold. Detecting a wake-up event may include detecting cardiac activity inconsistent with sleep, such as a change in heart rate variability indicative of a wake-up event, a heart rate above or below a predetermined threshold, or other changes in cardiac activity generally inconsistent with sleep. Detecting a wake-up event may also or alternatively include detecting a body temperature inconsistent with sleep, such as a body temperature above or below a predetermined threshold. As will be appreciated by those skilled in the art, other techniques and criteria may also or alternatively be used to detect a wake-up event. For example, a machine learning system may be trained to recognize a wake-up event based on any data available from the device or its environment. Such data includes external data such as the time of sunrise and the day of the week, as well as locally sensed data (e.g., an increase in light or a sudden sound in the user's surrounding area).

[0303] Regardless of the detection method, a wake-up event may be associated with the user's intent to go back to sleep or to remain awake. For example, in one aspect, the user may have the intent to wake up, which may be evidenced by subsequent activity and a persistent wakefulness. A wake-up event may alternatively include an event in which the user intends to go back to sleep, which may be evidenced by a return to a stationary prone position followed by sleep activity. These types of wake-up events may include, for example, general restlessness or the user's reaction to a sound disturbance. Therefore, it may be advantageous to determine the user's sleep intent before performing sleep analysis or other computationally expensive processing of the acquired physiological data.

[0304] As shown in step 1808, method 1800 may include evaluating the user's sleep intentions (e.g., whether the user intends to stay awake or go back to sleep) in response to the wake-up event.

[0305] The assessment of sleep intention may be based on past sleep data. For example, the past sleep data may be obtained from the user's sleep history or may be customized or adapted for a particular user. The past sleep data may also or instead include data obtained from other users, such as users of a platform or system for analyzing sleep or users with similar hardware or software for physiological monitoring. As will be appreciated, the past sleep data may be further filtered, categorized, weighted, or selected to improve the accuracy of the assessment of whether the user intends to stay awake or go back to sleep. For example, past sleep data of a population may be filtered to obtain data related to similarly situated users. This may be based on physiological, biological, or demographic data, such as age, height, weight, physical condition, diet, and medical history.

[0306] A user's intent may be assessed using a probabilistic analysis of past sleep data to calculate or estimate the probability that the user intends to stay awake or to go back to sleep according to one or more variables, including categorical and quantitative variables. For example, useful categorical variables may include day of the week, season, and / or geographic location. Useful quantitative variables may include the duration of the wake-up event, the length of the previous sleep interval, the amount of time offset from the usual wake-up time, the duration of post-wake movement, and / or the amount of time offset. Certain variables, such as time of day, may be used as either a quantitative variable (e.g., when calculating the time offset from the usual wake-up time) or a categorical variable (e.g., whether the wake-up time is within the usual wake-up interval), or both.

[0307] In one embodiment, thresholds may be applied to quantitative variables to derive rules for assessing sleep intention. For example, if the duration or amount of a user's physical movement after waking up exceeds a predetermined threshold for movement (which may be determined based on the user's past data or other past data, or may be a default threshold set by an administrator, etc.), it may be determined that the user intends to stay awake. In another embodiment, exceeding the threshold, or the amount by which the threshold is exceeded, may be used as evidence of an increased probability that the user intends to stay awake.

[0308] Variables used in the probabilistic analysis of past sleep data to determine the probability that a user intends to stay awake or intends to go back to sleep may be weighted. Such weights may be default weights for particular variables or may be customized weights for a particular user or set of users. For example, for a particular user with a sporadic sleep pattern, one or more quantitative variables may be weighted more heavily than one or more categorical variables. Certain variables, such as the length of the previous sleep interval or the amount or duration of movement after waking, may be good early indicators of sleep intent and may be used exclusively or in a heavily weighted manner to assess sleep intent.

[0309] In one embodiment, assessing the user's sleep intention may include training a machine learning algorithm to estimate a likelihood of a sleep intention outcome based on one or more characteristics of the previous sleep segment and / or wake-up event, such as any of the variables or sensed conditions described herein. The machine learning algorithm may be applied to the user's current data during the wake-up period to determine a probability distribution of sleep intention outcomes, and more generally, to determine whether the user is more likely to intend to wake up or to go back to sleep. Other probability models, such as probabilistic models, may also or instead be used to assess the likelihood that the user intends to stay awake.

[0310] As indicated at step 1810, method 1800 may include determining the user's sleep intent, i.e., as shown, determining whether the user intends to stay awake. This may include applying any of the above analyses, or any other suitable probability analysis, machine learning algorithm, rule-based evaluation, or other analysis, and combinations thereof, to determine whether the user is likely to intend to stay awake or to go back to sleep. In situations where the user is likely to request sleep metrics in the near future, it may be advantageous to perform the sleep intent assessment locally, such as on a wearable physiological monitor, to conserve network communication resources (including bandwidth and power consumption). To this end, it may be convenient to deploy a compact machine learning model or other algorithm on the wearable physiological monitoring device.

[0311] As shown in step 1812, method 1800 may include transmitting data from the physiological monitor to a remote server for sleep analysis if it is determined that the user intends to remain awake. However, it should be understood that the sleep analysis, or portions thereof, may also or instead be performed locally. For example, this may be performed using the processor and memory of the local computing device or the physiological monitoring device that acquired the data. This may include any form of preprocessing, signal conditioning, compression, encryption, or other data processing that may be useful to perform in a local computing environment before transmission to a remote resource. The sleep analysis may include any of the sleep analyses described herein or other sleep analyses known in the art. The sleep analysis may be usefully presented to the user on the local device as one or more sleep metrics assessing the quality and / or length of an overall sleep interval and / or individual sleep stages or sleep cycles within the overall sleep interval.

[0312] As shown in step 1814, if it is determined that the user intends to go back to sleep, method 1800 may include monitoring for additional wake-up events, where additional sensor data is acquired, e.g., by returning to step 1804, to assess the user's intent.

[0313] As will be appreciated, the above-described method 1800 may be implemented, in whole or in part, by a computer program product. For example, the computer program product may include computer-executable code embodied in a non-transitory computer-readable medium that, when executed on a wearable physiological monitor, performs the following steps: That is, acquiring sensor data including photoplethysmography data and accelerometer data by the wearable physiological monitor worn by the user during the sleep segment; detecting a wake-up event by the wearable physiological monitor by analyzing the accelerometer data to identify body movements inconsistent with sleep; in response to the wake-up event, using probability analysis based on past sleep data to evaluate whether the user intends to stay awake or to go back to sleep, and assessing the probability that the user intends to stay awake based on at least one of the length of the sleep segment and the duration of body movements inconsistent with sleep; if it is more likely that the user intends to stay awake, transmitting data from the wearable physiological monitor to a remote server for sleep analysis; and if it is more likely that the user intends to go back to sleep, monitoring for additional wake-up events and evaluating the user's intention when the additional wake-up events occur.

[0314] It will be understood that one or more of the steps associated with any of the methods described herein, or their associated sub-steps, calculations, functions, etc., may be performed locally, remotely, or a combination thereof. Thus, one or more of the steps of method 1800 of FIG. 18 may be performed locally on the wearable device, remotely on a server or other remote resource, on an intermediate device such as a local computer used by a user to access a remote resource, or any combination thereof. For example, when using example system 200 of FIG. 2 , one or more steps of a technique for detecting sleep intent may be performed, in whole or in part, locally on one or both of physiological monitor 206 and user device 220. For example, this may be performed by training a machine learning model to distinguish an intent to wake from an intent to go back to sleep, then lightweighting or optimizing the machine learning model and placing it on the wearable device, etc. Additionally or alternatively, one or more steps of the technique for detecting sleep intent may be implemented remotely, in whole or in part, in one or more of the remote server 230 and other resource(s) 250. Thus, for example, if a wearable monitor is positioned near the user's smartphone during sleep, heart rate data may be continuously or periodically transmitted to the remote server 230, which can monitor the received data to identify a possible or actual intent to wake up. Other combinations are also possible. For example, specific movement activity detected locally on the wearable device may be used as a trigger to cause the remote server 230 to begin evaluating sleep intent and may also be used to request updates of heart rate data, etc., as needed. Using any of these techniques, the remote server 230 can advantageously postpone computationally expensive sleep analysis until an intent to wake up is identified.

[0315] Stage-based coaching is now described, which may involve using physiological parameters measured over time from a wearable physiological monitoring device (e.g., respiration rate, heart rate variability, and temperature from any of the wearable sensors and devices described herein) to identify a reproductive or physiological stage, such as a pregnancy trimester, and adjust activity recommendations, such as sleep or exercise, accordingly.

[0316] It will be understood that any suitable physiological and / or hormonal stage may be detected and used to adjust activity recommendations as described herein. In some embodiments, reproductive stages, such as menstrual cycle stages, pregnancy trimesters, postpartum periods, menopause, and perimenopause, may be detected and / or measured to adjust recommendations to a user. Cyclical changes in human physiology may also respond to seasonal changes, length of daylight, weekly activity patterns, and the like. More generally, physiological rhythms are generally classified as ultradian (longer than one day), circadian (approximately one day), and subdian (less than one day). Such measurable or detectable rhythms, patterns, or periods may be used to synchronize and adjust recommendations to a user. Thus, unless expressly stated to the contrary or apparent from the context of this disclosure, where this disclosure uses particular stages as examples, it is understood that any suitable stages may also or instead be used.

[0317] Physiological rhythms observed during the reproductive phase can cause significant physiological changes, which may impact sleep, strain, and / or recovery. Concurrently, these rhythms may manifest as measurable changes in heart rate variability (HRV), resting heart rate (RHR), respiration, RR intervals, temperature, and more. This allows for automated detection and synchronized provision of recommendations related to an individual's sleep, strain, and / or recovery. For example, the early follicular phase of the menstrual cycle may be associated with greater strain during exercise compared to other phases of the menstrual cycle. Conversely, the late luteal phase of the menstrual cycle may be associated with advantageously reduced strain compared to other phases of the menstrual cycle, which may negatively impact recovery and the overall effectiveness of exercise. Therefore, detecting the menstrual cycle allows for automatic and flexible adjustment of activity recommendations to optimize a user's strain, recovery, and / or sleep, enabling them to achieve improved health and fitness goals.

[0318] Against this background, it would be advantageous to tailor activity recommendations to a user based on the identification of their reproductive phase. For example, if it is estimated that the follicular phase of a menstrual cycle requires less cardiovascular strain (compared to the luteal phase) to achieve the same next-day recovery level, the strain recommendations could take this into account, such as by encouraging more intense training earlier in the menstrual cycle. More generally, sleep, strain, and / or recovery coaching algorithms could adjust recommendations according to any reproductive phase, e.g., increase or decrease the recommended workout intensity based on the current phase.

[0319] In one embodiment, the reproductive stage may be automatically detected based on correlations between various stages and measurable physiological parameters (e.g., measured from a wearable physiological sensor worn substantially continuously by the user). However, the stage information may also or alternatively be manually entered by the user as a technique for the coaching algorithm to obtain the stage information or as a way to confirm the automatically detected stage. In general, the reproductive phase may be automatically detected based on resting heart rate, heart rate variability, respiratory rate, temperature, or a combination thereof. It will also be appreciated that the stage may be used to adjust recommendations regarding various activity plans, such as rest, sleep, nutrition, and exercise load.

[0320] 19 is a flow diagram illustrating a method 1900 for recommending activity plan adjustments based on reproductive stages. While this disclosure focuses on reproductive stages, it should be understood that method 1900 may also be applied to other suitable physiological and / or hormonal stages.

[0321] As shown in step 1902, method 1900 may include obtaining physiological data of a user from a wearable physiological monitoring device. The user may be a wearer of the wearable physiological monitoring device, which may be any one or more of the devices described herein.

[0322] The physiological data may include heart rate data, such as heart rate variability (HRV), resting heart rate (RHR), RR intervals, peak-to-peak data, and ECG data. The heart rate data may be obtained, at least in part, from one or more sensors in the wearable device using photoplethysmography (PPG). The physiological data may also or instead include other physiological data, such as data regarding oxygen levels, skin temperature, body temperature, sweat rate, and respiratory components. Furthermore, data from the wearable device may be processed to infer other physiological data, for example, by inferring respiration rate from heart rate variability. Furthermore, the data may include other data, such as summary data, motion data, fitness data, activity data, galvanic skin response data, or other data described herein or contemplated by one of ordinary skill in the art. In general, source data may be acquired at the wearable device, preprocessed in any manner, and transmitted to a remote resource for processing using any of the techniques described herein.

[0323] The physiological data may be collected substantially continuously by the wearable physiological monitoring device and stored on the wearable device until a suitable connection to a remote processing unit is available. Thus, acquiring the physiological data may include substantially continuously monitoring the user using a continuous physiological monitoring system and recording or calculating the physiological data as described herein. Continuous physiological monitoring has been described in considerable detail above and will not be repeated here. Generally, however, this may include raw sensor data, processed data, or a combination thereof, which may then be correlated with a reproductive cycle, such as a menstrual cycle. Additionally, method 1900 may include storing the physiological data in any raw or processed form on the wearable physiological monitoring device and / or transmitting the data to a local or remote location for storage, retrieval, and processing.

[0324] As shown in step 1904, method 1900 may include identifying the user's reproductive stage based on the physiological data. Reproductive stages may include one or more of the following: pregnancy trimester, postpartum period (e.g., the first three months after childbirth), menstrual cycle stage, menopausal stage, and perimenopausal stage. However, it should be understood that the reproductive stage may be any physiological stage based on reproductive hormones. In some embodiments, identifying the stage may include identifying the stage based on a pattern of change in the user's heart rate variability over a period of time. For example, in exemplary data obtained from over 3,000 wearers of physiological monitoring devices, heart rate variability (HRV) was, on average, approximately 8% higher than baseline during the early to mid-follicular phase of the menstrual cycle and approximately 4% lower than baseline during the mid to late luteal phase. Additionally, calculated recovery scores were approximately 10% higher during menstruation (i.e., the early follicular phase) and decreased to just above a 5% lower value than baseline during the luteal phase. Furthermore, when the relationship between strain and recovery the next day was examined in women with natural menstrual cycles, this phenomenon was found to persist even when strain was controlled for. However, the same phenomenon was not observed in women using contraceptives containing estrogen. In other words, this experimental data showed that wearers with natural menstrual cycles tended to have the highest recovery scores in the first week of their cycle. Specifically, when strain was controlled for, recovery scores in the early follicular phase were approximately 7 points higher (on average) than recovery scores in the early luteal phase.

[0325] Additionally or alternatively, identification of the user's reproductive stage may be based on a resting heart rate or resting heart rate pattern obtained from the user's heart rate data. Several techniques may be used to calculate the resting heart rate. The resting heart rate may be calculated, for example, once per day, based on measurements at a predetermined absolute time (e.g., 0100 hours) or relative time (e.g., the time immediately prior to waking). In another embodiment, a summary statistic, such as the mean, median, or minimum, may be calculated over a period of time and used as the resting heart rate. This may be during sleep or during one or more specific sleep stages, such as all slow-wave sleep periods. Alternatively, this may be calculated (e.g., averaged) over one or more periods of time during which motion or other data suggest the wearer is at rest during the day. Other techniques may also or alternatively be used.

[0326] In some embodiments, identifying the reproductive stage may include determining supplemental information about the reproductive stage. The supplemental information may include the duration of the stage, the onset of the stage, the probability that the identification information is accurate, etc. For example, if the identified reproductive stage is a trimester of pregnancy, the supplemental information may include the gestational age of the fetus.

[0327] Reproductive stage identification may use patterns of change in resting heart rate over time. This may involve simple pattern recognition or extraction of periodic characteristics of the changes (e.g., frequency analysis, etc.). In another embodiment, machine learning or statistical methods may be used to identify reproductive stages. For example, a machine learning model may be trained to identify reproductive stages based on one or both of the user's respiratory rate and resting heart rate. It will also be appreciated that reproductive stage identification may generally be fully automatic (e.g., using the techniques described herein), fully manual (e.g., based on the user's explicit report), or a combination thereof. In a semi-automatic mode, when a user reports a specific event, such as the onset of menstruation, transitions to other stages may be automatically determined based on a model derived from the user's previously manually entered stage history, population-level data, or a combination thereof. Additionally or alternatively, other user data, such as body mass index or age, may be relevant to stage prediction, and such data may be used to create physiologically similar cohorts (groups of people) for training.

[0328] Similarly, a user's respiration rate or patterns of change in their respiratory heart rate may be calculated using the heart rate variability of the user's heart rate data. For example, heart rate typically increases during inhalation and decreases during exhalation. By mapping this general phenomenon to continuous heart rate variability data via an algorithm, it is possible to determine when the user is inhaling and exhaling and derive the respiration rate. By calculating the daily respiration rate (e.g., as an average of the respiration rates measured that day, or as the respiration rate measured at a given absolute or relative time), patterns of change over time can be determined and used to identify the reproductive stage. In another embodiment, the respiration rate may be refined by first using peaks to identify the respiration pattern as described above, and then using frequency domain analysis to derive a second estimate of the respiration rate and identify peaks in the power spectrum (for the underlying time-based heart rate data) within a physiologically plausible range of respiration rate. The initial (time-based) estimate may be used to interpret the frequency domain estimate (which may show multiple plausible peaks), and the combined estimate (e.g., the frequency domain estimate closest to the time domain estimate) is reported as the current respiration rate. To avoid potentially misleading point estimates of respiratory rate, measurements may be averaged over a period of time (e.g., 1 or 2 minutes) using several overlapping window functions.

[0329] Continuous heart rate data can be used to identify reproductive stages, such as those described above, although other techniques may also or alternatively be used. For example, in one embodiment, reproductive stages may be identified based on user input, such as an explicit classification of one or more reproductive stages. In another embodiment, reproductive stages may be identified using physiological parameters, such as skin temperature, measured on the user (e.g., mapped to a history of changes in skin temperature over time using a wearable physiological monitoring device).

[0330] In one embodiment, identifying the reproductive stage may include training a machine learning model to detect the reproductive stage based on, for example, the user's respiratory rate and / or resting heart rate, or any of the other data sources described herein, as well as combinations thereof. That is, certain patterns may be known or may become known, and the machine learning model may be trained to identify these patterns and thereby identify the reproductive stage.

[0331] As shown in step 1906, method 1900 may include determining a user's current recovery level based on the user's past sleep activity. Various techniques are described herein for calculating an objective recovery score of the current recovery level based, for example, on the user's past sleep activity and past strain. However, other techniques for estimating physical recovery may also or instead be used. The current recovery level may be determined based on an estimate of strain using, for example, heart rate variability, resting heart rate, respiratory rate, etc.

[0332] The past sleep activity may be determined based on physiological data and / or other data, such as motion data, electroencephalogram data, eye movement data, body temperature data, and respiratory rate data, using any of the techniques described herein or any other techniques. Alternatively or additionally, the past sleep activity may be determined based on user input, such as reviewing the user's past sleep activity. The past sleep activity may include sleep duration for past sleep events (e.g., the previous night's sleep) and / or intermittent sleep activity (e.g., naps, etc.).

[0333] As shown in step 1908, method 1900 may include generating a recommended goal for the user's activity plan, for example, based on the user's current recovery level. Generally, the recommended goal may be determined based on data and / or indicators before taking into account the reproductive stage. In this case, the recommended goal may be determined in the same or similar manner as the calculation of the other activity plans described above. For example, the recommended goal may be determined at least in part based on sleep quality (e.g., as indicated by a sleep score described herein), recent physiological strain (e.g., as indicated by an intensity score described herein), heart rate data, activity data, combinations thereof, etc. The recommended goal may provide an objective measure of the activity plan, such as a calculated numerical goal or amount for the user. The activity plan may include any plan the user may undertake, such as a sport (e.g., football, soccer, golf, tennis, etc.), an exercise routine, a recreational activity (e.g., cycling, hiking, running, walking, meditation, etc.), sleep, rest, diet, hydration, etc. Thus, a recommended goal may be, for example, a calorie goal (e.g., 500 calories), a strain goal (understood to include recommendations related to the amount and / or intensity of activity, e.g., training volume and / or intensity, expressed using a strain score calculated in a numerical range of, e.g., 0 to 21 or other range), a power goal (e.g., distance, number of stairs, number of steps), or a sleep goal (e.g., sleep duration, sleep timing). Recommended goals may also or alternatively include activity goals such as a 20-minute run or a 30-minute swim. Recommended goals may also or alternatively include more user-specific composite recommendations, e.g., a 15-minute run at at least 8 miles per hour or a swim until 750 calories have been burned. More generally, any recommendation suitable for guiding a user to engage in an activity plan may be used as a recommended goal as contemplated herein.

[0334] As shown in step 1910, method 1900 may include automatically adjusting the user's activity plan by adjusting a recommended goal based on the reproductive stage. For example, if the reproductive stage is a phase of the menstrual cycle, this step 1910 may include automatically adjusting the user's recommended goal by increasing the activity plan intensity during the early follicular phase of the menstrual cycle and decreasing the activity plan intensity during the late luteal phase of the menstrual cycle. In another aspect, if the recommended goal is sleep duration, adjusting the recommended goal may include adjusting sleep duration. In another aspect, if the reproductive stage is a pregnancy trimester, this step 1910 may include automatically adjusting the user's recommended goal by increasing the user's dietary calorie goal during the second and third trimesters of pregnancy.

[0335] As shown in step 1912, method 1900 may include presenting the user with a recommended goal on a user interface. The recommended goal or other adjusted recommendation may be communicated to the user on the user interface in any of a variety of media, such as within a fitness application for a smartphone or other computing device associated with the wearable monitor or within a fitness website accessible to the user. The fitness website provides information and activity recommendations based on the user's data from the wearable device. In some embodiments, the user interface may present the user with a reproductive stage. Alternatively or additionally, the user interface may present supplemental information, such as alternative recommendations, predicted health and fitness changes that will occur if the recommended goal is achieved, a confidence level in the reproductive stage identification, a warning if the confidence level is below a predetermined threshold, and a reproductive stage duration.

[0336] 20A-20B illustrate correlations useful for automatically detecting menstrual cycle stages. As discussed above, the menstrual cycle stages may be used to recommend adjustments to activity plans. While the menstrual cycle is used for illustrative purposes, it should be understood that correlations can be detected for any suitable reproductive stage.

[0337] FIG. 20A shows a polynomial fit 2050 of actual resting heart rate data versus menstrual cycle days. Resting heart rate can be correlated with the menstrual cycle. Resting heart rate exhibits a pattern of variation over the course of the menstrual cycle, making it suitable for automated menstrual cycle detection. Continuing with the experimental data collected from the wearer of the physiological monitoring device described above, resting heart rate (RHR) decreased by approximately 5% during the mid-follicular phase and increased by approximately 3% during the luteal phase. This change in resting heart rate represents a significant change of nearly one standard deviation (when the data is normalized to the individual's baseline). Furthermore, respiratory rate was found to be lowest during the mid-follicular phase and highest during the luteal phase.

[0338] Figure 20B shows a polynomial fit 2060 of the respiration rate of the same group of users versus days in the menstrual cycle. Generally, respiration rate is lowest during the mid-follicular phase and highest during the luteal phase. Furthermore, this variation represents a change of approximately one standard deviation, making it suitable for automatically detecting the phase within the menstrual cycle.

[0339] 21-25, several examples of recommended activity plan adjustments are now described. While these examples relate to specific reproductive stages, similar techniques can be used for any other physiological stage that can be tracked manually and / or automatically, based on which activity plan adjustments can be advantageously made over the course of the stage. Furthermore, while the feedback in these examples generally relates to adjustments to health and fitness (e.g., exercise, diet, sleep, recovery, etc.), other feedback may also or instead be usefully provided.

[0340] 21 is a flow diagram illustrating a method for recommending workload-related adjustments to an activity plan based on a phase within a menstrual cycle. The user in this example may be a wearer of a wearable physiological monitoring device or a user of any physiological monitoring platform or system, such as those described herein. Method 2100 may use any of the data and data sources described herein.

[0341] As shown in step 2102, method 2100 may begin by identifying a stage within the user's menstrual cycle. The stage may be identified, for example, using any of the techniques described herein. For example, the stage may be identified using heart rate data (e.g., HRV, RHR, RR, etc.), respiration data (e.g., respiration rate, one or more oxygen levels, etc.), temperature data, etc., and / or the stage data may be received manually. The stage within the menstrual cycle may be identified as either the early follicular phase, the late follicular phase, the early luteal phase, or the late luteal phase. If the stage cannot be accurately identified, the user may be notified and, if necessary, queried for a clear identification of the stage. For each of the four stages identified in FIG. 21, activity recommendations for future exercise may be adjusted according to the current recovery level.

[0342] If the stage is identified as early follicular, method 2100 may include determining a current recovery level, as shown in step 2104. In this case, if the current recovery level is "low" and the user is in the early follicular phase of their menstrual cycle, method 2100 may include providing a recommendation for a medium level of strain to the user, as shown in step 2106. Also, under these conditions, if the current recovery level is "high" and the user is in the early follicular phase of their menstrual cycle, method 2100 may include recommending a high level of strain to the user, as shown in step 2108.

[0343] As shown in step 2110, method 2100 may include determining a current recovery level if the stage is identified as late follicular. In this case, as shown in step 2112, if the current recovery level is "low" and the user is in the late follicular phase of their menstrual cycle, method 2100 may include generating a low to medium strain recommendation for the user. As shown in step 2114, if the current recovery level is "high" and the user is in the late follicular phase of their menstrual cycle, method 2100 may include generating a medium to high strain recommendation for the user.

[0344] As shown in step 2116, method 2100 may include determining a current recovery level if the phase is identified as the early luteal phase. In this case, as shown in step 2118, if the current recovery level is "low" and the user is in the early luteal phase of their menstrual cycle, method 2100 may include generating a recommendation for a low level of strain for the user. As shown in step 2120, if the current recovery level is "high" and the user is in the early luteal phase of their menstrual cycle, method 2100 may include generating a recommendation for a medium level of strain for the user.

[0345] As shown in step 2122, method 2100 may include determining a current recovery level if the phase is identified as the late luteal phase. In this case, as shown in step 2124, if the current recovery level is "low" and the user is in the late luteal phase of their menstrual cycle, method 2100 may include generating a recommendation for a low level of strain for the user. As shown in step 2126, if the current recovery level is "high" and the user is in the late luteal phase of their menstrual cycle, method 2100 may include recommending a medium level of strain for the user.

[0346] 22 is a flow diagram illustrating a method for recommending fitness and nutritional adjustments to an activity plan based on the phase within a menstrual cycle. The user in this example may be a wearer of a wearable physiological monitoring device or a user of any physiological monitoring platform or system, such as those described herein. Method 2200 may use any of the data and data sources described herein. In general, guiding the type and intensity of exercise and the user's diet with various cycle-specific recommendations may improve the user's overall performance.

[0347] As shown in step 2202, method 2200 may begin by identifying a stage within the user's menstrual cycle. The stage may be identified, for example, using any of the techniques described herein. For example, the stage may be identified using heart rate data (e.g., HRV, RHR, RR, etc.), respiration data (e.g., respiration rate, one or more oxygen levels, etc.), temperature data, etc. Alternatively, the stage data may be received manually. The stage within the menstrual cycle may be identified as either the early follicular phase, the late follicular phase, the early luteal phase, or the late luteal phase. If the stage cannot be accurately identified, the user may be notified and, if necessary, queried for a clear identification of the stage.

[0348] As shown in step 2204, method 2200 may include generating recommendations to encourage high-intensity training if the stage is identified as the early follicular phase. This may include, for example, a general recommendation to engage in high-intensity activity and / or one or more explicit goals for intensity, such as goals for heart rate, distance, movement speed, weight use, calories, duration, etc. This may also or instead include explicit recommendations for exercise, including, for example, different activity types and intervals.

[0349] As shown in step 2206, method 2200 may include generating a recommendation for strength-based training and / or a longer warm-up time if the stage is identified as the late follicular phase. This recommendation may be generated as a measure to reduce injury to the user, as users may be more susceptible to injury during the late follicular phase of their menstrual cycle.

[0350] As shown in step 2208, method 2200 may include generating a recommendation to encourage low-intensity training if the phase is identified as the early luteal phase. As with other intensity-based recommendations, this may include recommendations for specific types, intensities, or durations of activity, or other objective measures of intensity, as well as combinations thereof. In another aspect, this may include a general recommendation to engage in low-intensity activity and / or a warning if the intensity exceeds a recommended range.

[0351] As shown in step 2210, method 2200 may include determining whether the user has recently completed an endurance-based sport or training session if the phase is identified as the late luteal phase. This determination may be made based on data from the sensors and / or data from user input.

[0352] As shown in step 2212, method 2200 may include providing dietary recommendations if the phase is identified as the late luteal phase and the user has recently completed an endurance-based sport or training session. For example, this may include recommending that the user consume a relatively large amount of carbohydrates as a way to replenish energy and speed recovery.

[0353] As shown in step 2214, if the phase is identified as the late luteal phase and the user has not recently completed an endurance-based sport or training session, method 2200 may include determining whether the user has recently completed a relatively high-intensity training session. This determination may be made based on data from sensors and / or user input.

[0354] As shown in step 2216, method 2200 may include providing one or more nutritional and fitness recommendations if the phase is identified as the late luteal phase and the user has recently completed a relatively high-intensity workout. For example, this may include recommending that the user hydrate before exercising and / or consume foods with a relatively high amount of sodium.

[0355] As shown in step 2218, method 2200 may include recommending to the user that they engage in relatively low intensity training if the phase is identified as the late luteal phase and the user has not recently completed relatively high intensity training.

[0356] 23 is a flow diagram illustrating a method for recommending sleep-related adjustments based on a stage within a menstrual cycle. The user in this example may be a wearer of a wearable physiological monitoring device or a user of any physiological monitoring platform or system such as those described herein. Method 2300 may use any of the data and data sources described above. In general, adjusting sleep recommendations for a user based on reproductive stage may improve overall user outcomes.

[0357] As shown in step 2302, method 2300 may begin by identifying a stage within a user's menstrual cycle. The stage may be identified, for example, using any of the techniques described herein. For example, the stage may be identified using heart rate data (e.g., HRV, RHR, RR, etc.), respiration data (e.g., respiration rate, one or more oxygen levels, etc.), temperature data, etc., or the stage data may be received manually. The stage within the menstrual cycle may be identified as either the early follicular phase, the late follicular phase, the early luteal phase, or the late luteal phase. If the stage cannot be accurately identified, the user may be notified and, if necessary, queried for a clear identification of the stage.

[0358] As shown in step 2304, method 2300 may include generating a recommendation to spend more time in bed if the stage is identified as the early follicular phase. Generating a recommendation to spend more time in bed may include recommending longer nighttime sleep, more frequent daytime naps, or an earlier bedtime, etc. However, it will be understood that any suitable sleep-related recommendation may be generated.

[0359] As shown in step 2306, method 2300 may include generating recommendations to spend less time in bed if the stage is identified as the late follicular phase. Generating recommendations to spend less time in bed may include recommendations to sleep less at night, take fewer daytime naps, go to bed later, etc. However, it will be understood that any suitable sleep-related recommendation may be generated.

[0360] As shown in step 2308, the method 2300 may include generating a recommendation to spend less time in bed if the phase is identified as the early luteal phase.

[0361] As shown in step 2310, the method 2300 may include generating a recommendation to spend more time in bed if the phase is identified as the late luteal phase.

[0362] While useful coaching recommendations and adjustments can be made according to hormonal cycles, such as the menstrual cycle, it will be appreciated that other human hormonal cycles may also or instead be used as the basis for adjusting recommendations regarding sleep (or other rest / restoration), exercise (or other activity), and / or nutrition. For example, reproductive stages, such as pregnancy and menopause, can have a significant impact on an individual's hormonal cycles and changes, the predictable physiological outcomes of which may be used as the basis for refining recommendations over the course of a hormonal cycle. Some examples are provided below:

[0363] 24 is a flow diagram illustrating a method for recommending sleep-related adjustments based on pregnancy trimester. The user in this example may be a wearer of a wearable physiological monitoring device or a user of any physiological monitoring platform or system, such as those described herein. Method 2400 may use any of the data and data sources described above. In general, adjusting sleep recommendations for a user based on reproductive stage may improve overall user outcomes.

[0364] As shown in step 2402, method 2400 may begin by identifying the user's hormonal stage, such as the trimester of pregnancy. Pregnancy can be detected automatically using continuous vital signs monitoring because pregnancy produces unique patterns in physiological indicators derived from inputs such as nighttime heart rate, heart rate variability, skin temperature, and pulse oximetry, as well as respiratory rate and sleep architecture. Combining these data, pregnancy can be reliably determined statistically between the fifth and sixth weeks of pregnancy. In addition to detecting the presence or absence of pregnancy, digital biomarkers that change as pregnancy progresses can also be used to identify pregnancy and determine the approximate gestational age of the fetus. Thus, the trimester can be identified using, for example, any of the techniques described herein. For example, the stage may be identified using heart rate data (e.g., HRV, RHR, RR, etc.), respiratory data (e.g., respiratory rate, one or more oxygen levels, etc.), temperature data, etc., and / or the stage data may be received manually. Alternatively or additionally, the stage may be identified from a user report (e.g., a home pregnancy test kit). In some embodiments, identifying the trimester of pregnancy may include identifying the gestational age of the fetus. If the trimester of pregnancy cannot be accurately identified, the user is notified and, if necessary, may be queried for a clear identification of the stage.

[0365] As shown in step 2404, if the stage is identified as the first trimester (early pregnancy), method 2400 may include generating a recommendation to stay in bed for a moderate amount of time. This may further include coaching recommendations, such as regular moderate-intensity aerobic exercise sufficient to raise the heart rate (but not exceeding about 140 beats per minute), brisk walking for 30 minutes at least five days per week, etc.

[0366] As shown in step 2406, if the stage is identified as the second trimester (mid-pregnancy), method 2400 may include generating a recommendation to encourage moderate to long periods of bed rest. This may additionally or alternatively include lowering the recovery score and / or reducing recommended workouts to achieve a particular user recovery score. This may additionally or alternatively include recommendations to reduce high-impact exercise that may increase the risk of injury due to ligament laxity, for example.

[0367] As shown in step 2408, method 2400 may include generating a recommendation to encourage moderate to long periods of bed rest if the stage is identified as the third trimester (late pregnancy). This may additionally or alternatively include lowering the recovery score and / or reducing recommended workouts for a particular user with a particular recovery score. Similarly, the strain score may be adjusted upward to reflect various aspects of increased strain due to pregnancy that may not be reflected in an HRV-based strain calculation. This may additionally or alternatively include a recommendation to reduce high-impact or very strenuous activities, such as long-distance running.

[0368] As shown in step 2410, method 2400 may include generating a recommendation to encourage moderate or increased time in bed if the stage is identified as the postpartum period (i.e., the "fourth" trimester). The recommendation may additionally or alternatively include reducing physical activity. Sleep scores may also be adjusted to encourage more rest and recovery. The postpartum period may be the three-month period after delivery, although it will be appreciated that any suitable period may be used instead.

[0369] 25 is a flow diagram illustrating a method for recommending sleep-related adjustments based on menopausal or perimenopausal stage. The user in this example may be a wearer of a wearable physiological monitoring device or a user of any physiological monitoring platform or system, such as those described herein. Method 2500 may use any of the data and data sources described herein.

[0370] As shown in step 2502, method 2500 may begin by identifying a user's hormonal stage, such as a menopausal stage or a perimenopausal stage. The stage may be identified, for example, using any of the techniques described herein. For example, the stage may be identified using heart rate data (e.g., HRV, RHR, RR, etc.), respiratory data (e.g., respiratory rate, one or more oxygen levels, etc.), temperature data, etc. Alternatively, the stage data may be received manually. Alternatively or additionally, the stage may be identified from a user report. In some embodiments, a menopausal stage may be identified by identifying a lack of menstrual cycles or a lack of menstrual cycle regularity. In some embodiments, a perimenopausal stage may be identified by identifying increased menstrual cycle irregularity. If the stage cannot be accurately identified, the user may be notified and, if necessary, queried for a clear identification of the stage.

[0371] If the stage is identified as a perimenopause stage, method 2500 may include determining a current recovery level, as shown in step 2504. In this case, if the current recovery level is "low," method 2500 may include providing a low to medium level of effort recommendation to the user, as shown in step 2506. Also, under these conditions, if the current recovery level is "high," method 2500 may include recommending a medium to high level of effort to the user, as shown in step 2508.

[0372] If the stage is identified as a menopausal stage, method 2500 may include determining a current recovery level, as shown in step 2510. In this case, if the current recovery level is "low," method 2500 may include providing a low level of effort recommendation to the user, as shown in step 2512. Also, under these conditions, if the current recovery level is "high," method 2500 may include recommending a medium level of effort to the user, as shown in step 2514.

[0373] FIG. 26 is a flow diagram of a method for providing coaching recommendations based on hormonal cycles. Generally, changes in hormone levels over the course of a hormonal cycle, such as a menstrual cycle, result in measurable changes in various physiological indicators that can be measured by a wearable monitor. By tracking these observable indicators over time and comparing them to expected values ​​for the cycle, a user's location within their cycle can be identified. As a key advantage, this monitoring can be performed automatically in the background, enabling appropriate coaching recommendations to be provided to the user in a timely manner.

[0374] As shown in step 2602, method 2600 may include providing a model of a hormonal cycle. In this context, providing a model may include storing the model where the model may be used in subsequent processing or creating the model. The model may be created, for example, by deriving a model hormonal cycle from a population of users, from a particular user's history, or a combination thereof. Generally, the model may characterize changes over time for each of several physiological indicators during the model hormonal cycle. For example, the physiological indicators may include heart rate variability, resting heart rate, body temperature, respiratory rate, etc. The hormonal cycle may include, for example, the user's menstrual cycle, the user's pregnancy, or the user's onset of menopause. While these cycles have different frequencies and durations, all of these cycles are intended to fall within the meaning of the term hormonal cycle as used herein.

[0375] The model may include any suitable data structure for estimating cycle position based on corresponding acquired physiological data. This may include, for example, a representation of changes in values ​​over time, which may facilitate comparing measured patterns to predicted patterns to identify where the user is in the relevant hormonal cycle. In another aspect, the data may be used to create machine learning, regression, or other models. These models allow for the calculation of predicted points in a hormonal cycle based on several measurements of underlying physiological indicators. This may include, for example, a regression or other predictive model for each physiological indicator of interest. These models can be used to generate predicted values ​​for comparison with measurements as the cycle progresses. In another aspect, analytical or empirical models can be created to generate time-varying explanations. The time-varying explanation can be compared to current measurements, or a series of current measurements can be compared to the time-varying explanation to identify the current point in the cycle. The period may be modeled, for example, using a composition of sinusoidal functions, as a dynamical system of differential equations, using machine learning models (such as recurrent neural networks, LSTM (long short-term memory) models), or any other model that can adequately represent a periodic repeating pattern.

[0376] It should also be understood that some hormonal cycles, while generally described as similar, have different characteristics, and different types of modeling and analysis may be appropriate for them. For example, a cycle such as pregnancy is a single-cycle phenomenon, and modeling it as a prediction of its end date may focus on estimating the duration until completion. Modeling may involve linear or time-series datasets, and the task may be to determine patterns or trends that help predict when an event will end. On the other hand, for recurring cycles such as the menstrual cycle, the relevant investigation is typically the current phase or position within a known recurring pattern. In this case, the expected length of the menstrual cycle may be less important than the current time or phase. Therefore, the analytical tools applied and / or models used may be different when determining timing within a pregnancy period than when determining timing within a menstrual cycle. At the same time, when physiological indicators provide timing information related to specific hormone levels, physiological states, etc., related coaching strategies may advantageously benefit from the modeling techniques described herein.

[0377] In general, physiological data may be sampled intermittently to ensure consistency and reliability of the data for subsequent use in modeling and analysis. For example, data may be measured once a day, e.g., at a specific time point, or at a specific time point within the user's sleep cycle. Additionally or alternatively, data may be an average of multiple measurements over a single day or over several days. For example, in one embodiment, multiple measurements may be taken during sleep, and the resulting stream of measurements may be weighted based on, e.g., recency, quality, sleep stage, etc., and calculated as a single derived measurement for that day. In another embodiment, an index may be created as a moving average over several days to smooth diurnal fluctuations in the data. More generally, any repeatable technique that facilitates comparison of multiple measurements may be used to obtain data for the applications contemplated herein.

[0378] As shown in step 2604, method 2600 may include acquiring data. This may include acquiring physiological data of the user from a wearable monitor, such as any of the monitors described herein. The physiological data may include, for example, heart rate data, body temperature data, etc., and may be acquired over the course of the user's hormonal cycle. In one embodiment, heart rate data may be readily acquired (e.g., if the wearable monitor is a photoplethysmograph) and converted to other physiological indices, such as resting heart rate, heart rate variability, respiratory rate, etc. In another embodiment, the wearable monitor may include a temperature monitor, and method 2600 may include acquiring temperature data from a temperature sensor and calculating the user's body temperature and / or skin temperature (at least daily or at the acquisition rate of other physiological indices).

[0379] As shown in step 2606, method 2600 may include calculating multiple metrics for the user at a predetermined interval. For example, this may include obtaining data at least daily during a hormonal cycle. Several factors unrelated to hormonal changes may cause temporal changes in one of the metrics. For example, an increase in body temperature due to illness or significant physical strain for one or more days may temporarily decrease heart rate variability. By obtaining multiple individually modeled metrics, such as two or more physiological metrics described herein, an ensemble approach may be advantageously used to improve detection accuracy and avoid undue influence of unrelated fluctuations in a single metric.

[0380] Computing multiple metrics may include, for example, calculating heart rate variability, resting heart rate, body temperature, and / or respiratory rate. In one embodiment, these metrics may be obtained directly from the wearable monitor, in which case little or no calculation may be required except for averaging, smoothing, filtering, etc. In another embodiment, the wearable monitor may provide a stream of raw pulse data, which can be converted by calculation into metrics of interest. In either case, method 2600 generally includes obtaining these physiological metrics for further processing as described herein.

[0381] As shown in step 2608, method 2600 includes calculating the user's estimated cycle time for the model hormone cycle based separately on each of the multiple (calculated) indicators. This may generally involve applying each of the physiological indicators obtained above to a model and calculating the corresponding cycle time suggested by that physiological indicator. The details of this calculation will depend on the nature of the model and may include, for example, applying a regression model to the data, identifying a match in a pattern over time (e.g., using any suitable time-domain and / or frequency-domain techniques), performing a lookup, or any other technique.

[0382] As shown in step 2610, method 2600 may include calculating a cycle time within a hormonal cycle based on the set of estimated cycle times. Various ensemble techniques are known in the art and can be used to process a group of estimated cycle times.

[0383] For example, in one embodiment, the set may include a weighted average of estimated cycle times for each of a plurality of indicators (e.g., physiological indicators or other indicators used to detect the timing of a hormonal cycle). The set may additionally or alternatively include a combination of estimated cycle times for each of the plurality of indicators based on the probability of accurately estimating cycle time. In another embodiment, the set may include a Bayesian model average of estimated cycle times, or an average of at least two of the estimated cycle times.

[0384] In another embodiment, rules may be applied to the ensemble. For example, predicted times may be withheld until two or more physiological indicators appear to agree on timing within a predetermined probability threshold (e.g., based on a probability estimator, etc.). In another embodiment, a single outlier indicator may be excluded if the remaining physiological indicators agree on the predicted timing. In this case, the user may be notified of the deviation from expectations and / or provided with relevant recommendations. More generally, various techniques are known for processing data from multiple sources, such as machine learning models and regression models. All such models are intended to fall within the scope of ensemble techniques, a term used herein.

[0385] As shown in step 2612, method 2600 may include providing coaching recommendations. Generally, this may include any of the coaching recommendations described herein. As a significant advantage, the recommendations may be provided based on a reliable determination of cycle time, and the recommendations may be tailored to a particular user if cycle time accelerates or slows during a particular cycle.

[0386] For example, increased progesterone during the luteal phase can lead to increased metabolism and hunger. In such situations, additional food intake may be recommended. As another example, carbohydrates are preferentially used for energy during the follicular phase, while fat is more readily utilized for energy during the luteal phase. Considering this, if weight loss is the goal, a leaner diet may be preferred during the luteal phase. More generally, to align macronutrient intake with metabolic trends, a carbohydrate-rich diet may be recommended during the follicular phase and a fat-rich diet may be recommended during the luteal phase. As another example, low-intensity training may be recommended during the luteal phase because carbohydrates are less readily utilized for energy and fatigue may occur more quickly. Conversely, the follicular phase is a favorable time for intense training because the ability to utilize carbohydrates as an energy source improves and pain sensitivity decreases. Therefore, intense strength training, etc., may be preferred during this phase. As another example, additional protein intake may be recommended during the late luteal phase because progesterone produced during this phase may increase catabolism.

[0387] More generally, any of the coaching strategies or recommendations described herein can be deployed in a manner synchronized with the timing of hormonal cycles calculated based on measured or calculated physiological indicators.

[0388] As mentioned above, in one aspect, the present specification discloses a wearable monitor, a model, and a processor configured to generate recommendations for a user. The wearable monitor may be configured to acquire heart rate data from a user. The model may be stored in memory and may characterize changes over time during a model hormone cycle for each of two or more physiological indicators. The processor may be configured to generate recommendations for a user by performing the following steps: receiving heart rate data from a wearable monitor; periodically calculating two or more physiological indicators for the user based on the heart rate data; calculating a cycle time within the user's hormone cycle based on a set of estimated cycle times, each estimated cycle time being derived by applying one of the physiological indicators to a model; and providing coaching information to the user based on the cycle time. In one aspect, the processor may execute on the user's personal computing device. In another aspect, the processor may execute on a remote server connected to the wearable monitor via a data network. In another aspect, the processor may include multiple processors distributed across these and / or other resources to perform these steps.

[0389] FIG. 27 illustrates a model of the menstrual cycle. Generally, model 2700 may include longitudinal data for multiple physiological indicators, such as heart rate variability, resting heart rate, skin temperature, and / or respiratory rate. These may be determined, for example, using the techniques described herein, based on historical data for a user or a group of users to establish consistency between measurements. As shown in FIG. 27, each physiological indicator has a characteristic shape of the mean value and characteristic variability over the course of a 28-day menstrual cycle. The characteristic variability may be a percentile range, standard deviation, or other variability indicator. This model may be used to predict timing, for example, by taking data and performing ensemble analysis to evaluate which point in the cycle a particular data set represents.

[0390] FIG. 28 illustrates a model of the pregnancy cycle. Typically, the model includes a separate longitudinal model for each physiological indicator, such as a resting heart rate model 2802, a heart rate variability model 2804, a respiration rate model 2806, and a skin temperature model 2808, each modeling the expected temporal change of the corresponding indicator based on time during pregnancy. The models may be based, for example, on a population of users, a specific user's history, or a combination thereof. Generally, pregnancy is divided into several distinct stages, such as (a) preconception, (b) first trimester (early pregnancy), (c) second trimester (mid-pregnancy), (d) third trimester (late pregnancy), and (e) postpartum. While the techniques described herein can be effectively used to identify which of these distinct stages a user falls into, using a set of multiple indicators allows for more accurate timing estimation, for example, by identifying days or weeks within the pregnancy. This can advantageously provide coaching recommendations that are better synchronized with the state of pregnancy and the user's specific hormone levels. This also makes it possible to identify conflicting trends in one or more metrics, which can be flagged to the user and recommended for further action.

[0391] 29 illustrates a method for detecting the onset of menopause. Generally, method 2900 may include providing a model as shown in step 2902, obtaining data as shown in step 2904, calculating an index as shown in step 2906, and monitoring hormone cycles as shown in step 2908, all as described, for example, with reference to method 2600 of FIG.

[0392] As shown in step 2910, method 2900 may include detecting the onset of menopause. In this context, various techniques may be employed to detect the onset of menopause. Generally, the onset of menopause is accompanied by changes in hormone levels associated with the menstrual cycle, resulting in corresponding changes (usually decreases) in the variability of physiological indicators related to hormone levels. At the same time, these changes may increase the variability in the frequency and duration of menstrual activity. This variability can be used to detect the onset of menopause and better tailor associated coaching recommendations to the accompanying physiological changes.

[0393] In one embodiment, detecting the onset of menopause may be based on the observed temporal irregularity of the hormone cycle. Thus, detecting the onset of menopause may include identifying one or more temporal irregularities of the hormone cycle compared to a model hormone cycle, calculating a probability indicating the onset of menopause from the one or more temporal irregularities, and providing a prediction of the onset of menopause to the user in response to calculating a probability of the onset of menopause that exceeds a predetermined threshold based on the one or more temporal irregularities. In one embodiment, identifying one or more temporal irregularities may include detecting a deviation from a model in at least one physiological indicator. In another embodiment, identifying temporal irregularities includes detecting a deviation in a set of two or more physiological indicators from a model. In another embodiment, identifying one or more temporal irregularities may include detecting a change in the expected length of the hormone cycle.

[0394] In another embodiment, detecting the onset of menopause may be based on a decrease in variability in a physiological indicator related to the user's hormone levels, in which case detecting the onset of menopause may include identifying a series of peaks within a hormone cycle for each of two or more physiological indicators, identifying a decrease in magnitude of each of the two or more physiological indicators over time for the series of peaks, and providing a prediction of the user's onset of menopause in response to the decrease in magnitude over time.

[0395] As shown in step 2912, method 2900 may include providing recommendations, such as any of the coaching or other recommendations described herein. As a non-limiting example, providing recommendations may include providing recommendations to the user based on the predicted onset of menopause. The recommendations may include at least one of dietary recommendations, sleep recommendations, and activity recommendations (e.g., exercise recommendations or non-exercise activity recommendations). In one embodiment, these recommendations may be tailored to, for example, the user's timing of menopausal onset to provide appropriate support and health guidance. In one embodiment, providing recommendations may include notifying the user of the predicted onset of menopause, for example, so that the user can consider appropriate measures.

[0396] In accordance with the above, disclosed herein is a system including a wearable monitor configured to acquire heart rate data from a user and a processor configured to perform the following steps: receiving the heart rate data from the wearable monitor; periodically calculating two or more physiological indicators of the user based on the heart rate data, the two or more physiological indicators having values ​​influenced by one or more hormones associated with the user's hormonal cycle; generating a predicted onset of menopause for the user based on a predetermined pattern of the two or more physiological indicators over time; and providing coaching information to the user based on the predicted onset of menopause.

[0397] In one embodiment, the hormonal cycle may be identified by applying the two or more physiological indicators to a hormonal cycle model, and the predetermined pattern may include one or more temporal irregularities in the hormonal cycle. In another embodiment, the hormonal cycle may be identified by applying the two or more physiological indicators to a hormonal cycle model, and the predetermined pattern may include a decrease in the magnitude of each of the two or more physiological indicators over time relative to a series of peaks within the hormonal cycle. In another embodiment, both predetermined patterns can be advantageously used together to more accurately assess the onset and timing of menopause.

[0398] The above-described systems, devices, methods, processes, etc. may be implemented in hardware, software, or any combination thereof suitable for the control, data acquisition, and data processing described herein. This includes implementation in one or more microprocessors, microcontrollers, embedded microcontrollers, programmable digital signal processors, or other programmable devices or processing circuits, and internal and / or external memories. Additionally or alternatively, this may include one or more application-specific integrated circuits, programmable gate arrays, programmable array logic components, or any other device(s) configured to process electronic signals. Furthermore, implementations of the above-described processes or devices may include computer-executable code written using a structured programming language such as C, an object-oriented programming language such as C++, or any other high- or low-level programming language (including assembly language, hardware description language, database programming language and technology). The computer-executable code may be stored, compiled, or interpreted for execution on any of the above-described devices. Implementations of the above-described processes or devices may also include heterogeneous combinations of processors, processor architectures, or combinations of different hardware and software.

[0399] Thus, in one aspect, each of the above methods, and combinations thereof, may be embodied in the form of computer-executable code that performs its steps when executed on one or more computing devices. In another aspect, the method may be embodied in the form of a system that performs its steps, may be distributed in various ways among multiple devices, or all of the functionality may be incorporated into a dedicated stand-alone device or other hardware. The code may be stored in a non-transitory manner in computer memory. The computer memory may be memory from which the program is executed (such as a random access memory associated with a processor), or may be a storage device such as a disk drive, flash memory, or other optical, electromagnetic, magnetic, infrared, or other device or combination of devices. In another aspect, any of the above systems and methods may be embodied in any suitable transmission or propagation medium that carries computer-executable code and / or any input or output from the computer-executable code. In another aspect, the means for performing the steps associated with the above processes may include any of the hardware and / or software described above. All such permutations and combinations are intended to be within the scope of the present disclosure.

[0400] The method steps of the various embodiments described herein, unless a different meaning is expressly given or apparent from the context, are intended to include any suitable manner of causing those method steps to be performed in a manner consistent with the patentability of the claimed invention(s) below. Thus, for example, performing step X may include any suitable manner of causing another party, such as a remote user, a remote processing resource (e.g., a server or cloud computer), or a machine, to perform step X. Similarly, performing steps X, Y, and Z may include any manner of directing or controlling any combination of such other individuals or resources to perform steps X, Y, and Z and obtain the benefit of those steps. Thus, unless a different meaning is expressly given or apparent from the context, the method steps of the various embodiments described herein are intended to include any suitable manner of causing one or more other parties or entities to perform those steps in a manner consistent with the patentability of the claimed invention(s) below. Such parties or entities need not be under the direction or control of any other party or entity, nor need they be located in any particular jurisdiction.

[0401] It will be understood that the above-described methods and systems are presented by way of example, and not by way of limitation. Numerous variations, additions, omissions, and other modifications will be apparent to those skilled in the art. Furthermore, the order of presentation of method steps in the above description and drawings is not intended to require that the described steps be performed in that order, unless a particular order is expressly required or apparent from the context. Thus, while specific embodiments have been shown and described, various changes and modifications in form and detail can be made without departing from the spirit and scope of the disclosure, as will be apparent to those skilled in the art, and are intended to form a part of the invention as defined in the following claims.

Claims

1. A computer program product comprising computer executable code embodied on a non-temporary computer-readable medium, wherein when the computer executable code is executed on one or more computing devices, The steps include providing a model that characterizes the temporal changes in heart rate variability, resting heart rate, body temperature, and respiratory rate during a model hormone cycle, and A step of acquiring physiological data of a user from a wearable monitor, wherein the physiological data includes at least heart rate data and body temperature data, and the physiological data is acquired during the user's hormonal cycle. A step of calculating a plurality of indicators of the user at least daily during the hormonal cycle, wherein the plurality of indicators include at least the heart rate variability, the resting heart rate, the body temperature, and the respiratory rate. A step of calculating a set of estimated cycle times for the user for the model hormone cycle, wherein each estimated cycle time in the set of estimated cycle times is an individual cycle time estimate based independently on one of the plurality of indicators; The steps include: calculating the probability that each individual period time estimate accurately estimates the index-dependent period time for one of the multiple indicators; A step of calculating the period time within the user's hormone cycle based on a combination of each individual period time estimate in the set of estimated period times and the probability that each individual period time estimate in the set of period times accurately estimates the index-dependent period time for one of the multiple indicators, A step of providing coaching information to the user based on the aforementioned cycle time. A computer program product that performs the following actions.

2. The computer program product according to claim 1, wherein the hormonal cycle includes the user's menstrual cycle.

3. The computer program product according to claim 1, wherein the hormonal cycle includes the user's pregnancy.

4. The computer program product according to claim 1, wherein the set includes a weighted average of the estimated period times for each of the plurality of indicators.

5. The computer program product according to claim 1, wherein the two or more physiological indicators include at least one of heart rate variability, resting heart rate, and respiratory rate.

6. We provide a model that characterizes the temporal changes of each of two or more physiological indicators during a model hormone cycle. The system acquires heart rate data from the wearable monitor worn by the user. Based on the heart rate data, calculate the two or more physiological indicators of the user at least every day, The system calculates a set of estimated cycle times for the user for the model hormone cycle, and each estimated cycle time in the set of estimated cycle times is an individual cycle time estimate based independently on one of the two or more physiological indicators. The probability that each individual period time estimate accurately estimates the index-dependent period time for one of the two or more physiological indicators is calculated. Based on the combination of each individual period time estimate in the set of estimated period times and the probability that each individual period time estimate accurately estimates the index-dependent period time for one of the two or more physiological indicators, the period time within the user's hormone cycle is calculated, and each estimated period time in the set is derived by applying one of the two or more physiological indicators to the model. Providing coaching information to the user based on the aforementioned cycle time. Methods that include...

7. The method according to claim 6, wherein the hormonal cycle includes the user's menstrual cycle.

8. The method according to claim 6, wherein the hormonal cycle includes the user's pregnancy.

9. The method according to claim 6, wherein the set includes a weighted average of the estimated period times for each of the physiological indicators.

10. The method according to claim 6, wherein the set includes combinations of estimated period times based on the probability of accurately estimating the period time.

11. The method according to claim 6, wherein the set includes the Bayesian model average of the estimated period time.

12. The method according to claim 6, wherein the set includes at least two averages of the estimated period times.

13. The method according to claim 6, wherein the wearable monitor includes a photoelectric volume plethysmography monitor.

14. The method according to claim 13, wherein the two or more physiological indicators include at least one of heart rate variability, resting heart rate, and respiratory rate.

15. The two or more physiological indicators mentioned above include body temperature, The wearable monitor includes a temperature sensor, The method according to claim 6, further comprising obtaining temperature data from the temperature sensor and calculating the user's body temperature at least daily.

16. The method according to claim 6, wherein the model hormone cycle is derived from a group of users.

17. The method according to claim 6, wherein the model hormone cycle is derived from the user's history.

18. A wearable monitor configured to acquire heart rate data from the user, A model stored in memory, which characterizes the temporal changes in each of two or more physiological indicators during the model hormone cycle, The step of receiving the heart rate data from the wearable monitor, A step of periodically calculating two or more physiological indicators of the user based on the heart rate data, A step of calculating a set of estimated cycle times for the user for the model hormone cycle, wherein each estimated cycle time in the set of estimated cycle times is an individual cycle time estimate based independently on one of the two or more indicators. A step of calculating the probability that each individual period time estimate accurately estimates the index-dependent period time based on one of the two or more physiological indicators. A step of calculating the period time within the user's hormone cycle based on a combination of each individual period time estimate in the set of estimated period times and the probability that each individual period time estimate in the set of period times accurately estimates the index-dependent period time for each of the two or more physiological indicators, wherein each of the estimated period times is derived by applying one of the two or more physiological indicators to the model, and The step of providing coaching information to the user based on the aforementioned cycle time. A processor configured to generate recommendations for the user by performing the following: A system that includes this.

19. The system according to claim 18, wherein the processor runs on the user's personal computing device.

20. The system according to claim 18, wherein the processor is executed on a remote server connected to the wearable monitor via a data network.

Images