Instantaneous Stress Algorithm for Wearable Computing Devices

Abstract

A method for monitoring a user's stress includes receiving multiple time-series data inputs from multiple biometric sensor electrodes of a wearable computing device. The time-series data inputs include continuous electrodermal activity data and at least one of heart rate data, skin temperature data, and heart rate variability data. The method also includes sequentially processing the time-series data inputs using multiple filtering techniques. The method further includes selecting a model from multiple models based on the type of data input received as the time-series data input, and applying the selected model to the processed time-series data input to calculate an index of the user's physiological response at a point in time. The selected model is thus adjusted to use all of the time-series data inputs in calculating the index of the physiological response. The method further includes controlling a function of the device when the index of the physiological response exceeds a threshold.

Description

[Technical Field]

[0001] The present disclosure relates generally to wearable computing devices, and more particularly to momentary stress algorithms for wearable computing devices. [Background technology]

[0002] Recent advances in technology, including technology available through consumer devices, have led to corresponding advances in health detection and monitoring. For example, biometric monitoring devices such as fitness trackers and smart watches can determine information about the pulse or motion of the person wearing the device.

[0003] Some biometric monitoring devices include various sensors for measuring multiple biological parameters that may be useful to the device's user, such as a heart rate sensor, a multipurpose electrical sensor for electrocardiogram (ECG) and electrodermal activity (EDA) applications, an infrared sensor, a gyroscope, an altimeter, an accelerometer, a temperature sensor, an ambient light sensor, Wi-Fi, GPS, a vibration sensor, a speaker, and a microphone, among others. As an example, some biometric monitoring devices use multipath electrical sensors to measure the EDA response of a user's skin. These responses are observed as sensitive electrical changes in skin conductance and are typically detected on the user's palm or fingertip using wet or dry electrode systems.

[0004] A typical EDA response can be measured on the palm or fingertip using at least two electrodes, and skin conductance is calculated using the measured electrical impedance. The EDA response is expressed as the phasic component of skin conductance, the skin conductance response (SCR), and is detected by identifying momentary spikes in skin conductance relative to the background isotonic measurement, the skin conductance level (SCL). Generally, SCR is more accurately observed from data collected from the user's palm or fingertip due to the higher sweat gland density in these areas.

[0005] Although SCR detection on the palm or fingertips has been comprehensively reported in the literature for assessing stress, SCL alone may be useful for assessing a user's stress. Measuring continuous electrodermal activity (cEDA) can be used to observe certain biological events, such as the body's response to an acute stress event. However, using cEDA measurements to assess acute stress events is difficult when using electrodes attached to the top surface of a biometric monitoring device because cEDA requires continuous skin contact to provide accurate readings (i.e., the user must position the skin over the electrode surface for the entire measurement period). Therefore, electrodes for cEDA are positioned on the bottom (skin-facing) surface of a biometric monitoring device to promote continuous contact with the skin and avoid the need for frequent user input to facilitate continuous EDA measurement.

[0006] More specifically, in terms of timing, the primary difference between SCL and SCR is that SCR occurs on the scale of seconds, while SCL is assessed over seconds, minutes, hours, and / or days. As an example, FIG. 1 shows a graphical representation of EDA amplitude versus time. As shown, the graph provides a comparison of the phasic skin conductance component (SCR), represented as a peak, with the isotonic skin conductance component (SCL). Thus, accurately detecting changes in SCL requires continuous measurement (over seconds / minutes / hours / days, etc.), and can be used to assess physiological stress, regardless of whether SCR is detected.

[0007] Therefore, a wearable computing device that continuously measures EDA for the purpose of accurately detecting momentary or acute stress events and displaying such events to the user would be a welcome development in the art. Additionally, input from other sensors (e.g., photoplethysmography data (such as amplitude), accelerometer data, etc.) may provide the user with additional context for assessing and / or displaying acute stress events. Summary of the Invention

[0008] Aspects and advantages of embodiments of the present disclosure will be set forth in part in the description that follows, or may be learned from the description, or may be learned by practice of the embodiments.

[0009] In one aspect, the present disclosure is directed to a method for monitoring a user's stress using a wearable computing device. The method includes receiving, via a processor communicatively coupled to the wearable computing device, multiple time-series data inputs from multiple biometric sensor electrodes of the wearable computing device. The multiple time-series data include the user's continuous electrodermal activity (cEDA) data and at least one of the user's heart rate data, the user's skin temperature data, and the user's heart rate variability (HRV) data. The method also includes sequentially processing the multiple time-series data inputs using multiple filtering techniques. Furthermore, the method includes selecting a model from multiple models based on the types of data inputs received as the multiple time-series data inputs. Furthermore, the method includes applying the selected model to the processed multiple time-series data inputs to calculate an index of the user's physiological response at a point in time, the selected model being adjusted to use all of the multiple time-series data inputs in calculating the index of the physiological response. Additionally, the method includes controlling a function of the wearable computing device when the index of the physiological response exceeds a threshold.

[0010] In another aspect, the present disclosure is directed to a wearable computing device. The wearable computing device includes an electronic display, a plurality of biometric sensor electrodes for sensing a plurality of time-series data inputs related to biometrics of a user of the wearable computing device, and at least one processor communicatively coupled to the plurality of biometric sensor electrodes. The processor(s) are configured to perform a plurality of operations, including, but not limited to, receiving a plurality of time-series data inputs. The plurality of time-series data includes a user's continuous electrodermal activity (cEDA) data and at least one of the user's heart rate data, the user's skin temperature data, and the user's heart rate variability (HRV) data. The operations further include processing the plurality of time-series data inputs in sequence using a plurality of filtering techniques; selecting a model from a plurality of models based on the types of data inputs received as the plurality of time-series data inputs; and applying the selected model to the processed plurality of time-series data inputs to calculate an index probability of a stress event by the user at a point in time. Thus, the selected model is adjusted to use all of the plurality of time-series data inputs in calculating the index probability of a stress event. Additionally, the actions include controlling a function of the wearable computing device when the indicator of the stress event exceeds a threshold.

[0011] These and other features, aspects, and advantages of various embodiments of the present disclosure will become better understood with reference to the following description and appended claims. The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the present disclosure and, together with the description, serve to explain associated principles.

[0012] Detailed descriptions of embodiments directed to those skilled in the art are set forth herein with reference to the accompanying drawings. [Brief explanation of the drawings]

[0013] [Figure 1]1 provides a graphical representation of electrodermal activity (EDA) amplitude (y-axis) versus time (x-axis) according to one embodiment of the present disclosure. [Figure 2] 1 provides a perspective view of a wearable computing device on a user's wrist, according to one embodiment of the present disclosure. [Figure 3] 1 provides a front perspective view of a wearable computing device according to one embodiment of the present disclosure. [Figure 4] 4 provides a rear perspective view of the wearable computing device of FIG. 3. [Figure 5] 4 provides an exploded view of the display of the wearable computing device of FIG. 3. [Figure 6] 1 provides a schematic diagram of an exemplary set of devices capable of communicating, according to one embodiment of the present disclosure. [Figure 7] 1 illustrates various controller components of an exemplary system that may be utilized in accordance with one embodiment of the present disclosure. [Figure 8] 1 illustrates a flow diagram of an embodiment of a method for monitoring indicators of a user's stress using a wearable computing device according to the present disclosure. [Figure 9] 1 provides a flowchart of an embodiment of a momentary stress algorithm for calculating an index of stress of a user of a wearable computing device at a point in time according to the present disclosure. [Figure 10] 1 illustrates an embodiment of a graphical representation of unfiltered cEDA data of a user of a wearable computing device during an exercise event, in accordance with the present disclosure. [Figure 11] 1 illustrates a graphical representation of an embodiment of a filtered cEDA of a user of a wearable computing device during an exercise event, in accordance with the present disclosure. [Figure 12]1 provides a schematic diagram of normalization coefficients being transferred from a backend to a mobile device and from the backend to a momentary stress algorithm application on a wearable computing device in accordance with the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0014] Reference will now be made in detail to the embodiments of the present invention, one or more examples of which are illustrated in the drawings. Each example is provided by way of explanation of the invention, and not limitation of the invention. Indeed, it will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the scope or spirit of the invention. For example, features illustrated or described as part of one embodiment can be used with other embodiments to yield still further embodiments. It is therefore intended that the present invention cover such modifications and variations as come within the scope of the appended claims and their equivalents.

[0015] overview Recent advances in technology, including technology available through consumer devices, have led to corresponding advances in health detection and monitoring. For example, devices such as fitness trackers and smart watches can determine information about the pulse or motion of the person wearing the device. However, the capabilities of traditional devices have limited the amount and type of health information that can be determined using such devices.

[0016] However, recent advances in miniaturization of sensors, electronics, and power sources have enabled personal health monitoring devices to be provided in sizes so small that they were previously impractical. For example, one biometric monitoring device includes a wristband with a housing that is approximately 1.6 inches wide, 1.6 inches long, and 0.5 inches thick. Such biometric monitoring devices typically include a display, battery, sensor, electronics package, wireless communication capability, power source, and interface buttons packaged within this small volume.

[0017] Additionally, some biometric monitoring devices include various sensors for measuring multiple biological parameters that may be beneficial to the device user, such as a heart rate sensor, a multipurpose electrical sensor for ECG and EDA applications, an infrared sensor, a gyroscope, an altimeter, an accelerometer, a temperature sensor, an ambient light sensor, Wi-Fi, GPS, a vibration sensor, a speaker, and a microphone, among others.

[0018] As an example, some biometric monitoring devices use multipath electrical sensors to measure the EDA response of a user's skin. These responses are observed as sensitive electrical changes in skin conductance and are typically detected on the palm or fingertip of the user's hand using wet or dry electrode systems. Therefore, the EDA response can be used to assess changes in the user's physiological stress.

[0019] A typical EDA response can be measured on the palm or fingertip using at least two electrodes, and skin conductance is calculated using the measured electrical impedance. The EDA response is expressed as the phasic component of skin conductance, the skin conductance response (SCR), and is detected by identifying momentary spikes in skin conductance relative to the background isotonic measurement, the skin conductance level (SCL). Generally, SCR is more accurately observed from data collected from the user's palm or fingertip due to the higher sweat gland density in these areas.

[0020] Although SCR detection on the palm or fingertips has been comprehensively reported in the literature for assessing stress, SCL alone may be useful for assessing a user's stress. By measuring continuous electrodermal activity (cEDA), SCL can be used to observe certain biological events, such as the body's response to acute physiological responses, such as stress events. However, using cEDA measurements to assess acute stress events is difficult when using electrodes attached to the top surface of a biometric monitoring device because cEDA requires continuous skin contact to provide accurate readings (i.e., the user must position the skin on the electrode surface for the entire measurement period). Therefore, electrodes for cEDA are positioned on the underside (skin-facing) surface of a biometric monitoring device to promote continuous contact with the skin and avoid the need for frequent user input to facilitate continuous EDA measurement.

[0021] Accordingly, the present disclosure is directed to a wearable computing device and a computer-implemented method for determining indicators of a user's stress at a point in time (e.g., probability of a stress event exceeding a threshold, stress intensity, stress type, etc.). For example, the wearable computing device may implement a Momentary Stress Algorithm (MSA) programmed into the wearable computing device, either as an on-device algorithm or an in-mobile application, that is used to determine indicators of stress events. In particular, the MSA may be configured to predict physical (or physiological) manifestations of stress at a particular time and send notifications of predicted stress events to the user. More specifically, in one embodiment, the MSA generally receives a combination of raw data inputs (e.g., heart rate data, cEDA data, skin temperature, acceleration, altimeter, and heart rate variability data) as time series data that is processed using various filtering techniques. Filtering techniques may generally involve excluding (e.g., removing) unwanted data, e.g., data sets, from the raw data inputs and / or modifying (e.g., enhancing) certain data, e.g., data and / or portions of certain data sets, from the raw data inputs. Sequentially filtering the raw data inputs may involve applying different filters accordingly, one after the other, each applied filter serving to filter out and / or modify the values ​​of the raw data inputs in an (algorithmically) predefined manner. For example, filtering techniques may include checking the raw data inputs for various modes of the device (e.g., sleep, exercise, uninterruptible mode, and / or off-wrist mode), such that data inputs collected during a particular one or all of these modes may be excluded from consideration.The device modes to be considered for exclusion may be selected by the user while using the wearable computing device and / or may be automatically associated with the raw input data by at least one processor of the wearable computing device (e.g., based on biometric sensor data).

[0022] For example, a user or processor(s) may switch a wearable computing device into a sleep mode, an exercise mode, an uninterruptible mode, or a non-wearable mode, thereby associating any input data collected during the respective mode with that mode. The raw input data collected during the corresponding mode may then be considered not representative of a physiological response, such as a stress event in general. Additionally, certain confounding factors may be used to filter the data input, and data input corresponding to extreme movement (or various other variables) may also be excluded. In one example, if the MSA detects increased movement accompanied by increased cEDA, the user may be exercising (and not stressed), and such data may be excluded from the stress calculation.

[0023] The final dataset may also be normalized, with additional imputations utilized as needed. In particular, in one embodiment, a mean and standard deviation (or any suitable normalization factor) may be required for each MSA data input at a time scale appropriate for that input. Additionally, for each input, various features may be calculated to convert the time series dataset into a single value. Once a single value is obtained for each input, the MSA is configured to determine which model to use to estimate the user's stress indicator(s). For example, the determination of the model to use may be based on which data inputs are available. Thus, in one embodiment, if all of the input data (e.g., heart rate data, cEDA data, skin temperature, and heart rate variability data, respectively) are available, the model selected is the model in which all input data are used to calculate the stress indicator(s). In contrast, if only one or two input datasets are available, the model selected is the model in which only the two input datasets are used to calculate the stress indicator(s). The selected model (which may be, for example, a logistic regression classifier) ​​may then be applied to the available data to determine the user's stress indicator(s). In further embodiments, the MSA may also include post-processing or smoothing of the user's stress indicator(s). Thus, the proposed solution includes (automatically) selecting a model for further evaluating the processed data based on the raw data input, e.g., based on the number of different input data types available or the number of time-series input data available. For example, in one embodiment, the MSA may require that a detected stress event be a certain length (such as about 3 to 5 minutes). In other embodiments, the MSA may group multiple stress events together if they occur within a certain proximity of each other (e.g., if the stress events are within 5 minutes of each other). Thus, in such an embodiment, the MSA concludes that the multiple stress events represent a common stress event, rather than multiple consecutive stress events.

[0024] Generally, functionality of the wearable computing device may be controlled when an indicator of a stress event exceeds a threshold. As outlined above, such functionality of the wearable computing device may be functionality of a display of the wearable computing device, e.g., resulting in the indicator of a stress event exceeding a threshold being displayed on the display. Alternatively or additionally, functionality of the wearable computing device controlled by the calculated indicator of a stress event exceeding a threshold may include generating and sending a stress event notification to, e.g., a user of the wearable computing device, and / or initiating one or more (software) applications on the wearable computing device, e.g., for mood logging, journaling, and / or participation recording, and / or triggering a user interaction process via the wearable computing device in which the user of the wearable computing device must actively acknowledge the notification of the stress event. Techniques and wearable computing devices may thereby be provided for more efficiently making a user aware of one or more potentially harmful stress events and for automatically providing, and in particular initiating, countermeasures to reduce the user's stress level.

[0025] Referring now to the drawings, exemplary embodiments of the present disclosure will be described in more detail.

[0026] Exemplary Devices and Systems Referring now to the drawings, FIGS. 2-5 illustrate perspective views of a wearable computing device 100 according to the present disclosure. In particular, as shown in FIG. 2, the wearable computing device 100 may be worn on a user's forearm 102 like a wristwatch. As such, as shown, the wearable computing device 100 may include a wristband 103 for securing the wearable computing device 100 to the user's forearm 102. Additionally, as shown in FIGS. 2, 3, and 5, the wearable computing device 100 has an exterior 105 and a housing 104 that contains electronics associated with the wearable computing device 100. For example, in one embodiment, the exterior 105 may be constructed of glass, polycarbonate, acrylic, or the like. Furthermore, as shown in FIGS. 2, 3, and 5, the wearable computing device 100 includes an electronic display 106 disposed within the housing 104 and viewable through the exterior 105. Additionally, as shown, wearable computing device 100 may also include one or more buttons 108 that may be implemented to provide a mechanism for activating various sensors of wearable computing device 100 to collect certain health data of the user. Additionally, in one embodiment, electronic display 106 may cover an electronic package (not shown), which may also be housed within housing 104.

[0027] 4 , the housing 104 of the wearable computing device 100 further includes a dorsal wrist side 110 configured to rest against the user's dorsal wrist when worn by the user, and a plurality of sensor electrodes 112 located on the dorsal wrist side 110 of the housing 104 to maintain skin contact with the user when worn on the user's wrist. Thus, in such embodiments, each of the sensor electrodes 112 continuously measures the user's electrical impedance at least at the location of skin contact on the dorsal wrist. Thus, in one or more embodiments, one or more (or all) of the plurality of sensor electrodes 112 may be cEDA sensor electrodes. In some embodiments, the wearable computing device 100 may also include at least one additional biometric sensor electrode in addition to the cEDA sensor electrode. In such embodiments, the additional biometric sensor electrode may include one or more temperature sensors (such as an ambient temperature sensor or a skin temperature sensor), humidity sensors, light sensors, pressure sensors, microphones, optical sensors, or photoplethysmography (PPG) sensors.

[0028] Additionally, the sensor electrodes 112 described herein may be constructed of any suitable material. For example, in one embodiment, the sensor electrodes 112 described herein may be constructed of stainless steel, graphene, or any other material with suitable electrical conductivity and / or corrosion resistance, and may have an optional PVD coating, which may be 1 micrometer thick titanium nitride. In such an embodiment, the PVD coating may impart a desired color to the sensor electrode 112, which may also provide oxidation protection and improved durability beyond what stainless steel already provides.

[0029] In additional embodiments, PVD and surface finishes can be used to increase / decrease water retention, which affects the cEDA signal and user comfort. In certain embodiments, the sensor electrode 112 may be formed from a tin and nickel alloy (TiN) with a shiny or mirror finish. Furthermore, in one embodiment, the sensor electrode 112 may be constructed from a hydrophobic or transparent material.

[0030] 6, components of an exemplary system 200 of the wearable computing device 100 that may be utilized in accordance with various embodiments are shown. In particular, as shown, the system 200 may also include at least one controller 202 communicatively coupled to the plurality of sensor electrodes 112. Additionally, in one embodiment, the controller(s) 202 may be a central processing unit (CPU) or a graphics processing unit (GPU) for executing instructions that may be stored in a memory device 204, such as flash memory or DRAM, among other options.

[0031] For example, in one embodiment, the memory device 204 may include RAM, ROM, FLASH memory, or other non-transitory digital data storage and may include a control program including sequences of instructions that, when loaded from the memory device 204 and executed using the controller(s) 202, cause the controller(s) 202 to perform the functions described herein. As will be apparent to one skilled in the art, the system 200 may include many types of memory, data storage, or computer-readable media, such as data storage for program instructions for execution by a controller or any suitable processor. The same or separate storage can be used for images or data, removable memory can be available for sharing information with other devices, and any number of communication approaches can be available for sharing with other devices.

[0032] Additionally, as shown, the system 200 includes any suitable display 206, such as a touchscreen, organic light emitting diode (OLED), or liquid crystal display (LCD), although the device may communicate information via other means, such as through audio speakers, a projector, or by casting to a display or streaming data to another device, such as a mobile phone, where an application on the mobile phone displays the data.

[0033] System 200 may also include one or more wireless components 212 operable to communicate with one or more electronic devices within communication range of a particular wireless channel. The wireless channel may be any suitable channel used to allow devices to communicate wirelessly, such as a Bluetooth, cellular, NFC, ultra-wideband (UWB), or Wi-Fi channel. It should be understood that system 200 may have one or more conventional wired communication connections known in the art.

[0034] System 200 also includes one or more power components 208, such as a battery operable to be recharged through a conventional plug-in approach or through other approaches, such as capacitive charging via proximity to a power mat or other such device. In further embodiments, system 200 may also include at least one additional I / O device 210 capable of receiving conventional input from a user. This conventional input may include, for example, push buttons, a touchpad, a touchscreen, a wheel, a joystick, a keyboard, a mouse, a keypad, or any other such device or element, by which a user can input commands into system 200. In other embodiments, I / O device(s) 210 may also be connected by a wireless infrared or Bluetooth or other link in some embodiments. In some embodiments, system 200 may also include a microphone or other audio capture element that receives voice or other audio commands. For example, in certain embodiments, system 200 may not include any buttons at all and may be controlled solely through a combination of visual and audio commands, such that a user can control wearable computing device 100 without having to make contact with the device. In an embodiment, the I / O elements 210 may also include one or more of the sensor electrodes 112 described herein, optical sensors, barometric pressure sensors (eg, altimeters, etc.), and the like.

[0035] 6 , system 200 may also include driver 214 and at least some combination of one or more emitters 216 and one or more detectors 218 (referred to herein as optical package 215) for measuring data for one or more metrics of a human body, such as a person wearing wearable computing device 100. In such an embodiment, as shown in FIG. 4 , for example, optical package 215 may be disposed within housing 104 and may be at least partially exposed through dorsal wrist side 110 of housing 104. Thus, as shown and further described herein, sensor electrodes 112 may be positioned around optical package 215 on wrist side 110 of housing 104. In alternative embodiments, various components of optical package 215 may be positioned around sensor electrodes 112 and / or in other suitable configurations, such as adjacent to, interspersed with, surrounded by, or above optical package 215. In some embodiments, for example where the sensor electrode 112 is transparent, the sensor electrode 112 may be disposed over the optical package 215 .

[0036] In some embodiments, system 200 may include at least one imaging element, such as one or more cameras, capable of capturing images of the surrounding environment and capable of imaging a user, people, or objects near the device. The imaging element may include any suitable technology, such as a CCD image capture element having sufficient resolution, focusing range, and viewing area to capture images of a user as the user operates the device. Additional image capture elements may also include depth sensors. Methods for capturing images using a camera element with a computing device are well known in the art and will not be described in detail herein. It should be understood that image capture may be performed using a single image, multiple images, periodic imaging, continuous image capture, image streaming, etc. Additionally, system 200 may include the ability to start and / or stop image capture, such as upon receiving a command from a user, an application, or another device.

[0037] The emitter 216 and detector 218 of FIG. 6 may also be used to obtain optical PPG measurements, in one example. Some PPG techniques rely on detecting light at a single spatial location, adding signals obtained from two or more spatial locations, or an algorithmic combination thereof. Both of these approaches result in a single spatial measurement from which a heart rate (HR) estimate (or other physiological metric) can be determined. In some embodiments, the PPG device uses a single light source (i.e., a single optical path) coupled to a single detector. Alternatively, the PPG device may use multiple light sources (i.e., two or more optical paths) coupled to a single detector or multiple detectors. In other embodiments, the PPG device uses multiple detectors coupled to a single light source or multiple light sources (i.e., two or more optical paths). In some cases, the light source(s) may be configured to emit one or more of green, red, infrared (IR) light, and any other suitable wavelengths in the spectrum (e.g., long IR for metabolic monitoring). For example, the PPG device may use a single light source and two or more photodetectors, each configured to detect a specific wavelength or range of wavelengths. In some cases, each detector is configured to detect a different wavelength or wavelength range from the others. In other cases, two or more detectors are configured to detect the same wavelength or wavelength range. In still other cases, one or more detectors are configured to detect a particular wavelength or wavelength range that is different from one or more other detectors. In embodiments using multiple optical paths, the PPG device may determine an average of the signals resulting from the multiple optical paths before determining an HR estimate or other physiological metric.

[0038] Additionally, in one embodiment, emitter 216 and detector 218 may be coupled directly or indirectly to controller 202 using driver circuits that enable controller 202 to drive emitter 216 and obtain signals from detector 218. Host computer 222 may communicate with wireless network component 212 via one or more networks 220, which may include one or more local area networks, wide area networks, UWB, and / or internetworks using either terrestrial or satellite links. In some embodiments, host computer 222 executes control and / or application programs configured to perform some of the functions described herein.

[0039] Referring now to FIG. 7 , a schematic diagram of an environment 300 in which aspects of various embodiments may be implemented is shown. In particular, as shown, a user may have several different devices capable of communicating using at least one wireless communication protocol. For example, as shown, the user may have a smartwatch 302 or a fitness tracker (such as wearable computing device 100), and the user desires that the smartwatch 302 or fitness tracker be able to communicate with a smartphone 304 and a tablet computer 306. The ability to communicate with multiple devices may enable the user to use applications installed on either the smartphone 304 or the tablet computer 306 to obtain information from the smartwatch 302, such as data captured using sensors on the smartwatch 302. The user may also desire that the smartwatch 302 be able to communicate with a service provider 308, or other such entity, that can obtain and process data from the smartwatch and provide functionality that may not otherwise be available on the applications installed on the smartwatch or individual devices. Additionally, as shown, the smartwatch 302 may be able to communicate with a service provider 308 through at least one network 220, such as the Internet or a cellular network, or may communicate with one of the individual devices via a wireless connection, such as Bluetooth, which in turn may communicate via at least one network. In various embodiments, there may be some other type of communication or reason for communication.

[0040] In addition to being able to communicate, users may also want devices to be able to communicate in some way or manner. For example, users may want communications between devices to be secure, especially if the data may include personal health data or other such communications. Device or application providers may also need to protect this information, at least in some circumstances. Users may want devices to be able to communicate with each other simultaneously, rather than sequentially. This may be especially true when pairing may be required, as users may prefer each device to be paired at most once so that manual pairing is not necessary. Users may also want communications to be as standards-based as possible, so that little manual intervention is required on the user's part, but also so that devices can communicate with as many other types of devices as possible, often outside of various proprietary formats. Thus, users may want to be able to walk around a room with one device and have such device automatically communicate with other target devices with little or no effort on the user's part. In various conventional approaches, devices utilize communication technologies such as Wi-Fi to communicate with other devices using wireless local area networks (WLANs). Smaller or lower volume devices, such as many Internet of Things (IoT) devices, instead utilize communication technologies such as Bluetooth®, particularly Bluetooth Low Energy (BLE), which consumes very little power.

[0041] 7 allows data to be captured, processed, and displayed in several different ways. For example, data may be captured using sensors on the smartwatch 302, but due to limited resources on the smartwatch 302, the data may be transferred to the smartphone 304 or service provider 308 (or cloud resources) for processing, and the results of that processing may then be presented to the user on the smartwatch 302, smartphone 304, and / or other such device associated with the user, such as a tablet computer 306. In at least some embodiments, the user may also be able to provide input, such as health data, using an interface on any of these devices, which may then be considered in making the decision.

[0042] Referring now to FIG. 8 , a flow diagram of one embodiment of a method 400 for monitoring a user's stress using a wearable computing device is provided. In one embodiment, for example, the wearable computing device may be any suitable wearable computing device, such as the wearable computing device 100 described herein with reference to FIGS. 1-7 . Accordingly, the method 400 is generally described herein with reference to the wearable computing device 100 of FIGS. 1-7 . However, it should be understood that the disclosed method 400 may be performed on any other suitable wearable computing device having any other suitable configuration. In addition, while FIG. 8 depicts steps performed in a particular order for purposes of illustration and description, the methods described herein are not limited to any particular order or arrangement. Using the disclosure provided herein, one skilled in the art will understand that various steps of the methods disclosed herein may be omitted, rearranged, combined, added, and / or adapted in various manners without departing from the scope of the present disclosure.

[0043] As mentioned above and described herein, the wearable computing device includes a plurality of biometric sensor electrodes on a dorsal wrist side of a housing of the wearable computing device. Accordingly, as shown at (402), method 400 includes receiving a plurality of time-series data inputs from the plurality of biometric sensor electrodes of wearable computing device 100 via a processor communicatively coupled to wearable computing device 100. In such an embodiment, by way of example, the plurality of time-series data may include a user's cEDA data, as well as a user's heart rate data, a user's skin temperature data, a user's heart rate variability (HRV) data, accelerometer data, altimeter data, and / or combinations thereof. In particular embodiments, for example, the plurality of time-series data inputs may include a user's heart rate, a user's skin temperature, a user's heart rate variability, and a user's cEDA data, and the combination of such data inputs results in an improved estimation of the user's stress.

[0044] As shown at (404), method 400 includes processing the multiple time series data inputs in sequence using multiple filtering techniques. For example, as described in more detail herein, processing the multiple time series data inputs in sequence using multiple filtering techniques may include filtering the user's cEDA data using, for example, a high pass filter for SCR, a low pass filter for SCL, a median filter to eliminate glitches, and / or any other type of filter as desired.

[0045] Further, in one embodiment, processing the plurality of time series data inputs in sequence using the plurality of filtering techniques includes updating a particular time frame cache with the plurality of data inputs (i.e., updating a cache that stores data inputs associated with a time frame / time window of a predetermined length); determining whether a time series data input from the plurality of time series data inputs is indicative of one of a plurality of modes of the wearable computing device associated with undesired motion (e.g., from exercise), and if so, excluding the time series data input from the plurality of time series data inputs; and filtering the plurality of time series data inputs based on the plurality of confounding factors to exclude one or more of the plurality of confounding factors from the plurality of time series data inputs. filtering out or modifying time series data inputs that satisfy one or more of the following criteria: imputing one or more data points into the plurality of time series data inputs if a minimum number of data points are missing from the plurality of time series data inputs or discarding the plurality of time series data inputs if a certain number of data points are missing from the plurality of time series data inputs; normalizing the plurality of time series data inputs using one or more normalization factors, wherein the one or more normalization factors comprise at least one of a mean, a median, a mode, or a standard deviation; and / or processing the plurality of time series data inputs in sequence using a plurality of filtering techniques may include converting each of the plurality of time series data inputs to a single value.

[0046] 8 , as shown at (406), method 400 includes selecting a model from a plurality of models based on the types or values ​​of data inputs received as the plurality of time-series data inputs. As shown at (408), method 400 includes applying the selected model to the processed plurality of time-series data inputs to calculate an indicator of a physiological response, such as a stress event, of a user at a point in time by the user, the selected model being adjusted to use all of the plurality of time-series data inputs in calculating the indicator of the stress event.

[0047] As shown at (410), method 400 includes providing a user with an indication of stress events at a time via a display. More specifically, in one embodiment, method 400 may include sending a notification to the user indicating at least one of an indication of stress events, a graphical representation of stress events over time, and / or a summary of stress events over time (e.g., daily, weekly, monthly). Further, in one embodiment, method 400 may also include surveying the user to respond to the notification via a display (e.g., display 206). For example, the user may be prompted to log their mood, journal, record their participation in a prescribed stress relief activity (e.g., meditation, walking, listening to music, guided breathing), and / or any other appropriate response.

[0048] Generally, a function of the wearable computing device may be controlled when an indicator of a stress event exceeds a threshold. As outlined above, such a function of the wearable computing device may be a function of the display 206 of the wearable computing device 100, for example, resulting in the indicator of the stress event exceeding a threshold being displayed on the display 206. Alternatively or additionally, a function of the wearable computing device 100 controlled by the calculated indicator of the stress event exceeding a threshold may include generating and sending a stress event notification to, for example, the user of the wearable computing device, and / or initiating one or more (software) applications on the wearable computing device 100, for example, for mood logging, journaling, and / or participation recording, and / or triggering a user interaction process via the wearable computing device 100 in which the user of the wearable computing device 100 must actively acknowledge the notification of the stress event. This may provide techniques and wearable computing devices 100 to more efficiently make a user aware of one or more potentially harmful stress events and automatically suggest or even initiate countermeasures to reduce the user's stress level.

[0049] Method 400 of Figure 8 may be better understood with respect to Figures 9-11. More specifically, Figure 9 shows a flowchart of an embodiment of an instantaneous stress algorithm 500 for calculating an index of a stress event of a user of a wearable computing device at a point in time, according to the present disclosure. In particular, as shown at (502), algorithm 500 includes incorporating on-device data from wearable computing device 100. As mentioned above, such raw data may include, for example, the user's cEDA data, as well as the user's heart rate data, the user's skin temperature data, accelerometer data, altimeter data, and / or the user's HRV data, and combinations thereof.

[0050] As shown at (504), algorithm 500 includes calculating a minute current signal for a portion of the raw data (e.g., heart rate data, HRV data, and / or cEDA data). More specifically, in one embodiment, as shown at (506), algorithm 500 may include applying a filter to the cEDA data, e.g., using a high-pass filter for SCR, a low-pass filter for SCL, a median filter to eliminate glitches, and / or any other type of filter as desired. For example, as shown in FIGS. 10 and 11 , graphical representations 600, 700, respectively, of embodiments of a wearable computing device user's cEDA data (e.g., SCL, y-axis) versus time during an exercise event (x-axis) according to the present disclosure are shown. However, FIG. 10 shows raw data, while FIG. 11 shows filtered data, e.g., via a high-pass filter. Further, as shown in FIG. 10 , SCL data is represented at 602, cEDA event data, such as a stress event, is represented at 604, and exercise detection is represented at 606. 11, the SCL data is represented by 702, the cEDA / SCL data is represented by 704, and the high-pass filtered data is represented by 706 and 708, respectively. In particular, as shown, the difference between line 706 and line 708 is the cutoff frequency used to generate 706 vs. 708, with 708 having a lower cutoff frequency (e.g., 120 min), meaning that the change is maintained over a longer time interval. In other words, line 706 returns to zero approximately six times faster than line 708.

[0051] 10 and 11, the cEDA / SCL data signal 602 may decay very slowly (e.g., over several hours). Therefore, the algorithm 500 can filter out exercise events from the cEDA data without losing all the predicted power on the "downslope" side of the peak, which may contain useful data. However, if linear interpolation is used between the onset of exercise and the "downslope" side of the peak, the algorithm 500 essentially reintroduces exercise data. Therefore, to address this issue, a high-pass filter can be applied to the cEDA signal 602, for example, after the data has passed through a slew rate limiter, to avoid introducing abrupt changes to the value or slope before filtering the data. Because such changes are preserved by filtering, they lead to abrupt spikes appearing in the filtered value. In a simplified sense, the algorithm monitors for changes in conditions where abrupt spikes would cause many false predictions of stress events. The outputs, represented by 706 and 708, are the filtered cEDA data.

[0052] If raw data is not available (or not enough raw data is available), then algorithm 500 terminates at (526) since a prediction is not possible. However, if raw data is available, algorithm 500 continues at (508). In particular, as shown at (508), algorithm 500 updates the cache for a certain time window (e.g., an X-minute window) for one or more of the data inputs. Furthermore, in one embodiment, as shown at (510), algorithm 500 may receive a set window length (e.g., 30 minutes) for updating the cache. It should be understood that the window length can be selected as any suitable time frame and is not limited to 30 minutes.

[0053] Further, as shown at (512), algorithm 500 includes determining whether the time series data input from the raw data indicates one of a plurality of modes of wearable computing device 100 related to motion. In such an embodiment, for example, the modes of wearable computing device 100 may include a sleep mode, an exercise mode, or an unworn mode. If yes, no prediction is possible, and algorithm 500 terminates at (526). However, if no, algorithm 500 continues at (514).

[0054] In particular, as shown at (514), when wearable computing device 100 is not operating in one of the aforementioned modes, algorithm 500 is configured to filter the raw data to remove a certain number of raw data based on a plurality of confounding factors, and to exclude or modify raw data that meets one or more of the plurality of confounding factors. In such an embodiment, the plurality of confounding factors may include, for example, an increase in the user's cEDA data with increased motion (as generally shown by graphical representation 600 in FIG. 10 ) as shown at (516), a percentage of HRV data above a certain threshold as shown at (520), a certain confidence level of the user's heart rate data as shown at (518), wearable computing device 100 being partially or fully submerged in liquid detectable using a combination of cEDA and altimeter data, skin contact between wearable computing device 100 and / or the user based on how many sensor electrodes 112 are simultaneously in contact with the user's skin per minute, or humidity above a humidity threshold.

[0055] 9 , as shown at (522), algorithm 500 is further configured to either impute one or more data points into the raw data if a certain number of data points are missing from the raw data, or to discard the raw data if the number of missing data points exceeds a threshold. Thus, as shown, if the number of missing data points exceeds a threshold, algorithm 500 terminates at (526) because no prediction is possible. However, if the raw data is missing sufficiently few values / data points, algorithm 500 imputes all missing data points, for example, using interpolation or extrapolation, and continues at (528).

[0056] In particular, as shown at (528), algorithm 500 is further configured to normalize the raw data using one or more normalization factors. In such an embodiment, for example, the normalization factor(s) may include the mean, median, mode, standard deviation, or any other suitable statistical function over a time period appropriate for each input. For example, as shown in FIG. 12, the normalization factor(s) may be transported via file transfer, similar to how settings may be transferred to wearable computing device 100. In the illustrated embodiment, for example, transport mechanism 800 requests backend 802 to expose an HTTP endpoint 804 that can be queried by companion application 808, e.g., on mobile device 806. Companion application 806 can then query these values ​​at predefined time intervals, and any changes in the payload will result in companion application 808 sending the normalization factor to wearable computing device 100. The wearable computing device 100 can decrypt the file, persistently store the new normalization weights, and immediately apply them to the algorithm 500, for example, via the MSA algorithm application 812 in user space 810.

[0057] 9 , algorithm 500 is configured to extract certain features from raw data by applying several commonly used time series transformation functions, as shown at (530). In particular, algorithm 500 is configured to receive (or may be programmed with) certain features and / or hyperparameters that can be used to extract certain features from raw data, as shown at (532). In one embodiment, for example, algorithm 500 is configured to convert each of the time series raw data inputs into a single value. Thus, algorithm 500 is configured to apply certain time series transformation functions independently to each of the input signals.

[0058] Thus, as shown at (534), algorithm 500 is then configured to determine / select a model from a plurality of models based on the types and / or values ​​of data inputs received in the raw data. For example, as shown at (536), if the raw data includes heart rate data, HRV data, cEDA data, and skin temperature, the selected model is configured to use all of the received raw data to determine an indicator of a stress event experienced by the user at a given time. In other embodiments, if there is little raw data available (e.g., cEDA data and skin temperature), algorithm 500 is configured to select a model that uses a subset of the available data types (e.g., cEDA data and skin temperature) to determine an indicator of a stress event experienced by the user at a given time. It should be understood that the algorithm can therefore select a model tailored to the available raw data.

[0059] Thus, as shown at (538), algorithm 500 is configured to apply the selected model to the data to generate an output that may be representative of a physical (or physiological) manifestation of the user's stress. In a further embodiment, by way of example, the selected model may be a machine learning model. For example, as shown at (540), the machine learning model may be a logistic regression model, a deep neural network, or any other suitable machine learning model now known in the art or later developed.

[0060] As shown at (542), algorithm 500 is also configured to post-process the indicators of the user's stress events. In such embodiments, for example, post-processing the indicators of the user's stress events may include ensuring that the stress events have durations exceeding a certain threshold. In other embodiments, post-processing the indicators of the user's stress events may include grouping multiple stress events together if they occur within a certain time frame of each other. Algorithm 500 ends at (544).

[0061] Thus, as described above, the output of the algorithm is an indication of stress events experienced by the user at a point in time. Thus, in further embodiments, wearable computing device 100 may send the user a notification indicating the indication of the stress event, a graphical display of the stress events over time, and / or a summary of the stress events over time. Further, as described above, wearable computing device 100 may poll the user to respond to the notification via the display, such as by requesting the user to participate in mood logging, journaling, and recording participation in prescribed stress relief activities.

[0062] Additional Disclosures In addition to the above, the systems, programs, or features described herein may provide users with controls that allow them to choose both if and when collection of user information (e.g., information about the user's social network, social actions or activities, occupation, user preferences, or the user's current location) may be enabled, and whether content or communications are sent from the server to the user. Furthermore, certain data may be processed in one or more ways so that personally identifiable information is removed before it is stored or used. For example, the user's identity may be processed so that personally identifiable information about the user cannot be determined, or if location information is obtained (e.g., to the city, zip code, or state level), the user's geographic location may be generalized so that the user's specific location cannot be determined. Thus, users may control what information is collected about them, how that information is used, and what information is provided to them. To that end, any information collected as described herein related to the user will be kept private and confidential and will not be used or disclosed inappropriately.

[0063] The technology described herein refers to servers, databases, software applications, and other computer-based systems, as well as actions performed on and information sent to and from such systems. The inherent flexibility of computer-based systems allows for a wide variety of possible configurations, combinations, and divisions of tasks and functionality among components. For example, the processes described herein can be implemented using a single device or component, or multiple devices or components working in combination. Databases and applications can be implemented on a single system or distributed across multiple systems. Distributed components can operate sequentially or in parallel.

[0064] While the present subject matter has been described in detail with respect to various specific exemplary embodiments thereof, each example is provided by way of explanation and not as a limitation of the present disclosure. Those skilled in the art, once they arrive at the foregoing understanding, will be able to readily create modifications, variations, and equivalents to such embodiments. Accordingly, the disclosure of the subject matter does not exclude the inclusion of such modifications, variations, and / or additions to the subject matter as would be readily apparent to one skilled in the art. For example, features illustrated or described as part of one embodiment may be used with other embodiments to yield still further embodiments. Accordingly, the present disclosure is intended to cover such modifications, variations, and equivalents.

Claims

1. 1. A method for monitoring a user's stress using a wearable computing device, comprising: receiving, via a processor communicatively coupled to the wearable computing device, a plurality of time-series data inputs from a plurality of biometric sensor electrodes of the wearable computing device, the plurality of time-series data including continuous electrodermal activity (cEDA) data of the user, and at least one of heart rate data of the user, skin temperature data of the user, and heart rate variability (HRV) data of the user, the method further comprising: processing the plurality of time series data inputs in sequence using a plurality of filtering techniques; selecting a model from a plurality of models based on types of data input received as the plurality of time series data inputs; applying the selected model to the processed plurality of time series data inputs to calculate an index of the physiological response of the user at a point in time, wherein the selected model is adjusted to use all of the plurality of time series data inputs in the calculation of the index of the physiological response, the method further comprising: controlling a function of the wearable computing device when the indicator of the physiological response exceeds a threshold.

2. The method of claim 1 , wherein the plurality of time series data inputs comprises the user's heart rate, the user's skin temperature, the user's heart rate variability, and the user's cEDA data.

3. processing the plurality of time series data inputs in sequence using the plurality of filtering techniques, The method of claim 1 or 2, further comprising filtering the cEDA data of the user using a high-pass filter, a low-pass filter, or a median filter.

4. processing the plurality of time series data inputs in sequence using the plurality of filtering techniques, The method of any one of claims 1 to 3, further comprising updating a particular time frame cache with said plurality of data entries.

5. processing the plurality of time series data inputs in sequence using the plurality of filtering techniques, 10. The method of claim 1, further comprising: determining whether a time series data entry from the plurality of time series data inputs indicates one of a plurality of modes of the wearable computing device, and if so, excluding or modifying the time series data entry from the plurality of time series data inputs.

6. The method of claim 5 , wherein the plurality of modes of the wearable computing device include one of a sleep mode, an exercise mode, an uninterruptible mode, or a non-worn mode.

7. processing the plurality of time series data inputs in sequence using the plurality of filtering techniques, 10. The method of any one of the preceding claims, further comprising filtering the plurality of time series data inputs based on a plurality of confounding factors to exclude or modify time series data inputs from the plurality of time series data inputs that satisfy one or more of the plurality of confounding factors.

8. 8. The method of claim 7, wherein the plurality of confounding factors include one of: an increase in the cEDA data of the user with increasing motion; a percentage of the HRV data above a certain threshold; a certain confidence level of the heart rate data of the user; a motion classifier based on accelerometer values; the wearable computing device being partially or fully submerged in or exposed to liquid; skin contact between the wearable computing device and the user being below a contact threshold; or humidity being above a humidity threshold.

9. Processing the plurality of time series data inputs using the plurality of filtering techniques includes:

10. The method of any one of the preceding claims, further comprising: imputing one or more data points into the plurality of time series data inputs if a certain number of data points are missing from the plurality of time series data inputs; or discarding the plurality of time series data inputs if the number of missing data points exceeds a threshold.

10. Processing the plurality of time series data inputs using the plurality of filtering techniques includes:

10. The method of any one of the preceding claims, further comprising normalizing the plurality of time series data inputs using one or more normalization factors.

11. 10. The method of any one of the preceding claims, wherein the selected model is a machine learning model.

12. Processing the plurality of time series data inputs using the plurality of filtering techniques includes:

10. The method of claim 1, further comprising converting each of the plurality of time series data inputs to a single value.

13. further comprising post-processing the indicator of the physiological response; 10. The method of any one of the preceding claims, wherein post-processing the indicators of the physiological responses further comprises at least one of ensuring that the physiological responses include a duration above a certain threshold, and grouping multiple physiological responses together if they occur within a certain time frame of each other.

14. 10. The method of claim 1, wherein controlling the function of the wearable computing device includes at least one of controlling a display of the wearable computing device and providing the indication of the physiological response at the time to the user via the display.

15. 15. The method of claim 14, further comprising sending a notification to the user via the display indicating at least one of the occurrence of the indicator of the physiological response exceeding a threshold, a graphical representation of the physiological response over time, and a summary of the physiological response over time.

16. 16. The method of claim 14 or 15, further comprising prompting the user to respond to the notification via the display of the wearable computing device, wherein responses to the notification include at least one of logging mood, journaling, guided breathing, guided meditation, and recording participation in a prescribed stress relief activity.

17. 10. The method of claim 1, wherein the processor is part of one of the wearable computing device or a separate mobile device.

18. 1. A wearable computing device, comprising: An electronic display, a plurality of biometric sensor electrodes for sensing a plurality of time-series data inputs related to biometrics of a user of the wearable computing device; at least one processor communicatively coupled to the plurality of biometric sensor electrodes, the at least one processor configured to perform a plurality of operations, the plurality of operations including: receiving the plurality of time-series data inputs, the plurality of time-series data including continuous electrodermal activity (cEDA) data of the user, and at least one of heart rate data of the user, skin temperature data of the user, and heart rate variability (HRV) data of the user, the plurality of operations further comprising: processing the plurality of time series data inputs in sequence using a plurality of filtering techniques; selecting a model from a plurality of models based on types of data input received as the plurality of time series data inputs; applying the selected model to the processed plurality of time series data inputs to calculate an index probability of a stress event for the user at a point in time by the user, wherein the selected model is adjusted to use all of the plurality of time series data inputs in the calculation of the index probability of the stress event, and the plurality of operations further comprise: and controlling a function of the wearable computing device when the indicator of the stress event exceeds a threshold.

19. processing the plurality of time series data inputs in sequence using the plurality of filtering techniques, filtering the cEDA data of the user using a high-pass filter, a low-pass filter, or a median filter; updating a particular time frame cache with the plurality of data entries; determining whether a time series data entry from the plurality of time series data entry indicates one of a plurality of modes of the wearable computing device, and if so, excluding or modifying the time series data entry from the plurality of time series data entry; filtering the plurality of time series data inputs based on a plurality of confounding factors to exclude or modify time series data inputs from the plurality of time series data inputs that satisfy one or more of the plurality of confounding factors; imputing one or more data points into the plurality of time series data inputs if a certain number of data points are missing from the plurality of time series data inputs, or discarding the plurality of time series data inputs if the number of missing data points exceeds a threshold; normalizing the plurality of time series data inputs using one or more normalization factors, wherein the one or more normalization factors comprise at least one of a mean, a median, a mode, or a standard deviation for a time scale appropriate to each time series data input; and processing the plurality of time series data inputs in sequence using the plurality of filtering techniques further comprises: The wearable computing device of claim 18 , further comprising converting each of the plurality of time-series data inputs into a single value.

20. further comprising post-processing the indicator probability of the stress event; 20. The wearable computing device of claim 18 or 19, wherein post-processing the indicator probability of the stress event further includes at least one of ensuring that the stress event includes a duration above a certain threshold, and grouping multiple stress events together if they occur within a certain time frame of each other.

21. sending a notification to the user via the electronic display indicating at least one of the occurrence of the indicator of the stress event exceeding a threshold, a graphical representation of the stress event over time, and a summary of the stress event over time; and surveying the user to respond to the notification via the electronic display; 21. The wearable computing device of claim 18, wherein responses to the notification include at least one of mood logging, journaling, guided breathing, guided meditation, and recording participation in a prescribed stress relief activity.

22. 22. The wearable computing device of claim 18, wherein controlling the function of the wearable computing device includes at least one of controlling the electronic display of the wearable computing device and providing the indication of the stress event at the time to the user via the electronic display.

Images