Fall risk assessment for users

Abstract

A computer-implemented method is provided for assessing a user's risk of falling. The computer-implemented method includes obtaining data indicative of the user engaging in a fall risk activity. The computer-implemented method includes adjusting a fall detection threshold of a wearable computing device worn by the user based at least in part on the data indicative of the user engaging in the fall risk activity. The method includes providing a notification indicating that the user is at risk of falling due, at least in part, to the user's engagement in the fall risk activity.

Description

[Technical field]

[0001] (Priority Claim) This application claims priority to U.S. patent application Ser. No. 63 / 295,046, entitled "Fall Risk Assessment for a User Wearing a Wearable Computing Device," filed Dec. 30, 2021, which is incorporated herein by reference.

[0002] The present disclosure relates generally to wearable computing devices and, more particularly, to methods for assessing a risk of falling for a user wearing one or more wearable computing devices. [Background technology]

[0003] The wearable computing device may be worn, for example, on a user's wrist. The wearable computing device may include one or more sensors (e.g., accelerometers, gyroscopes, etc.) that may obtain data indicative of a user's movement. In some examples, data obtained from the one or more sensors may indicate the occurrence of a fall event in which a user falls to the ground. In some examples, the wearable computing device may alert emergency personnel in response to detecting the fall event. Summary of the Invention

[0004] Aspects and advantages of embodiments of the present disclosure are set forth in part in the detailed description that follows, or may be learned from the detailed description or through practice of the embodiments.

[0005] In one aspect, a computer-implemented method is provided for assessing a user's risk of falling. The method includes obtaining data indicative of the user engaging in a fall risk activity. The method includes adjusting a fall detection threshold of a wearable computing device worn by the user based at least in part on the data indicative of the user engaging in the fall risk activity. The method includes providing a notification indicating that the user is at risk of falling due, at least in part, to the user's engagement in the fall risk activity.

[0006] In some implementations, the method includes obtaining a signal from a mobile computing device associated with the user. The signal can indicate that the user is engaged in a fall risk activity. Further, in such implementations, the method includes providing a notification for display on a display screen of the mobile computing device.

[0007] In some implementations, obtaining data indicative of fall risk activity includes obtaining data from one or more sensors of a wearable computing device worn by the user.

[0008] In some implementations, adjusting the fall detection threshold of the wearable computing device includes providing data from one or more sensors of the wearable computing device as input to a machine-learned model configured to process the data from the one or more sensors of the wearable computing device to determine an adjusted fall detection threshold. Further, adjusting the fall detection threshold of the wearable computing device includes obtaining the adjusted fall detection threshold as an output of the machine-learned model.

[0009] In some implementations, the one or more sensors of the wearable computing device include a plurality of different sensors, and the adjusted fall detection threshold includes a plurality of adjusted fall detection thresholds, further, each of the plurality of adjusted fall detection thresholds is associated with a respective sensor of the plurality of different sensors.

[0010] In some implementations, the fall risk activity includes walking after waking from a sleep event. Further, in such implementations, the data indicative of the user engaging in the fall risk activity includes movement data indicative of the user walking at above a threshold speed immediately after waking from a sleep event.

[0011] In some implementations, the computer-implemented method includes acquiring sleep data associated with the sleep events. Further, in such implementations, the computer-implemented method includes adjusting a fall-detection threshold of the wearable computing device according to the sleep data and the movement data.

[0012] In some implementations, the sleep data includes at least one of a duration of a sleep event or a depth of a sleep event. Additionally, the sleep data indicative of the depth of a sleep event indicates whether the user is aroused from a rapid eye movement (REM) sleep cycle or aroused from a non-REM sleep cycle.

[0013] In some implementations, obtaining data indicative of the user engaging in a fall risk activity includes obtaining first data from one or more sensors of a first wearable computing device worn at a first location on the user's body and obtaining second data from one or more sensors of a second wearable computing device worn at a second location on the user's body. The second location on the body may be different from the first location on the body. Further, in such implementations, adjusting the fall detection threshold of the wearable computing devices includes adjusting a fall detection threshold of at least one of the first wearable computing device or the second wearable computing device.

[0014] In another aspect, a wearable computing device is provided. The wearable computing device includes a plurality of sensors. The wearable computing device further includes one or more processors communicatively coupled to the plurality of sensors. The one or more processors are configured to obtain data indicative of a user wearing the wearable computing device engaging in a fall risk activity. The one or more processors are further configured to adjust a fall detection threshold of the wearable computing device based at least in part on the data indicative of the user's engagement in the fall risk activity. The one or more processors are further configured to provide a notification indicating that the user is at risk of falling due, at least in part, to the user's engagement in the fall risk activity.

[0015] In yet another aspect, a non-transitory computer-readable medium is provided having instructions stored thereon that, when executed by one or more processors, cause the one or more processors to perform operations. The operations include obtaining data indicative of a user wearing a wearable computing device engaging in a fall risk activity. The operations include adjusting a fall detection threshold of the wearable computing device based at least in part on the data indicative of the user engaging in the fall risk activity. The operations include providing a notification indicating that a user engaging in the fall risk activity is at risk of falling because, at least in part, the user is engaged in the fall risk activity.

[0016] These and other features, aspects, and advantages of various embodiments of the present disclosure will become better understood with reference to the following detailed description and the 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 detailed description, serve to explain associated principles.

[0017] A detailed discussion of embodiments directed to those skilled in the art is set forth herein with reference to the accompanying drawings. [Brief description of the drawings]

[0018] [Figure 1] 1 illustrates a user wearing multiple wearable computing devices according to some embodiments of the present disclosure. [Diagram 2] 1 illustrates a network for communication between wearable computing devices worn by users, according to some embodiments of the present disclosure. [Diagram 3] FIG. 1 illustrates a block diagram of components of a wearable computing device according to some embodiments of the present disclosure. [Figure 4] FIG. 1 illustrates a block diagram of a machine-learned model for adjusting a fall detection threshold of a wearable computing device according to some embodiments of the present disclosure. [Diagram 5] 1 shows a flow diagram of a method for assessing fall risk according to some embodiments of the present disclosure. [Figure 6] 1 illustrates a flow diagram of a method for adjusting a fall detection threshold according to some embodiments of the present disclosure. [Figure 7] 1 shows a flow diagram of a method for assessing risk of falling after awakening from a sleep event according to some embodiments of the present disclosure. [Figure 8] FIG. 1 illustrates a front perspective view of a wearable computing device according to some embodiments of the present disclosure. [Figure 9] 1 illustrates a rear perspective view of a wearable computing device according to some embodiments of the present disclosure. [Figure 10] 1 illustrates a computing system according to some embodiments of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] Reference will now be made in detail to the embodiments of the present disclosure, one or more examples of which are illustrated in the drawings. Each example is provided as an explanation of the disclosure, and not as a limitation of the disclosure. Indeed, it will be apparent to those skilled in the art that various modifications and variations can be made in the present disclosure without departing from the scope or spirit of the disclosure. For example, features illustrated or described as part of one embodiment can be used with another embodiment to yield still a further embodiment. 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.

[0020] Exemplary aspects of the present disclosure are directed to a wearable computing device that may be worn, for example, on a user's wrist. The wearable computing device may include one or more sensors (e.g., accelerometer, gyroscope, etc.) configured to acquire data indicative of a user's movements. The wearable computing device may generate a notification (e.g., visual, audible, etc.) in response to the one or more sensors acquiring data indicative of movements corresponding to a fall event. For example, the notification may include a text notification displayed on a display screen of the wearable computing device. Additionally, in some examples, the wearable computing device may notify emergency personnel if the user does not take any action (e.g., dismiss the notification).

[0021] An exemplary aspect of the present disclosure is directed to fall risk prevention for a user wearing a wearable computing device. The fall risk prevention may include detecting that a user is engaged in a fall risk activity. The fall risk prevention further includes adjusting a fall detection threshold in response to detecting that the user is engaged in a fall risk activity. For example, the fall detection threshold may be adjusted (e.g., lowered) so that fall detection is more sensitive while the user is engaged in the fall risk activity. Furthermore, the fall risk prevention further includes notifying the user that the risk of falling has increased due at least in part to the user being engaged in the fall risk activity. As used herein, a "fall detection threshold" may refer to a fall indication parameter output by one or more fall detection devices. Furthermore, a fall event may be detected when the fall indication parameter exceeds the fall detection threshold. It should be understood that the one or more fall detection devices may include any suitable device configured to detect a user's movement indicative of a fall event or a movement indicative of an impending fall event. For example, in some implementations, the one or more fall detection devices may include one or more motion sensors (e.g., accelerometer, gyroscope). Alternatively, or in addition, the one or more fall detection devices may include a barometer.

[0022] In some implementations, the fall risk activity may include walking while looking at a display of a mobile computing device (e.g., a smartphone). In such implementations, the mobile computing device may detect that the user is engaged in the fall risk activity and may communicate data indicating that the user is engaged in the fall risk activity to the wearable computing device. Additionally, the wearable computing device may adjust (e.g., lower) the fall detection threshold according to the fall risk activity in which the user is engaged. For example, in response to determining that the user is engaged in a first fall risk activity, the wearable computing device may lower the fall detection threshold by a first amount. Conversely, in response to determining that the user is engaged in a second fall risk activity that is different from the first fall risk activity, the wearable computing device may lower the fall detection threshold by a second amount that is different from the first amount.

[0023] In some implementations, the wearable computing device can notify a user that the user is at increased risk of falling due to engaging in a fall risk activity. For example, the wearable computing device can generate a notification (e.g., audible, visual) indicating the user's risk of falling while engaging in a fall risk activity. In some implementations, the notification can be communicated to a mobile computing device and displayed on a display screen of the mobile computing device.

[0024] In some implementations, the fall risk activity may include carrying an object (e.g., a box, a drink, etc.) that requires the user to use both hands to support the object. For example, one or more sensors (e.g., accelerometers) may acquire motion data indicating that the user is walking without swinging their arms. In this manner, the wearable computing device may determine that the user is engaged in a fall risk activity (e.g., carrying an object). Further, the fall detection threshold may be adjusted (e.g., lowered) while the motion data indicates that the user is engaged in an activity that involves a risk of falling.

[0025] In some implementations, the wearable computing device can generate a notification indicating that the user is engaging in a fall risk activity (e.g., carrying an object). For example, in some implementations, the wearable computing device can include a haptic feedback device (e.g., a piezoelectric buzzer) configured to provide haptic feedback (e.g., vibration) to prompt the user to put the object down and read the notification indicating the fall risk activity provided on a display of the wearable computing device. In an alternative implementation, the wearable computing device can emit an audible notification (e.g., a computer-generated voice) indicating the fall risk activity.

[0026] In some implementations, the fall risk activity may include walking after waking from a sleep event. The risk of falling within the first few steps after waking from a sleep event may vary based on various parameters (e.g., time spent sleeping, depth of sleep, movement before taking the first few steps, pace of the first few steps). For example, a user who walks immediately after waking from a sleep event may be more susceptible to a fall event compared to a user who waits a period of time before walking after waking from a sleep event.

[0027] As another example, the duration of a sleep event may affect the likelihood of a fall event occurring within the first few steps after waking from a sleep event. For example, a user waking from a sleep event lasting several hours may be more susceptible to a fall event compared to a user waking from a sleep event of shorter duration (e.g., one hour or less). The depth of a sleep event may also affect the likelihood of a user falling within the first few steps after waking from a sleep event. For example, a user waking from a deeper sleep event (e.g., including one or more rapid eye movement (REM) sleep cycles) may be more susceptible to a fall event compared to a user waking from a lighter sleep event (e.g., including non-REM sleep cycles).

[0028] As yet another example, the pace at which a user walks after waking from a sleep event may affect fall risk. For example, if a user walks quickly (e.g., above a threshold cadence) immediately after waking from a sleep event, the user may be more susceptible to a fall event. Conversely, if a user walks slowly (e.g., below a threshold cadence) immediately after waking from a sleep event, the user may be less susceptible to a fall event.

[0029] In some implementations, the fall detection threshold can be adjusted (e.g., lowered) at least in part according to one or more of the above-mentioned parameters. For example, the fall detection threshold can be lower for a user awakened from a deep sleep event than for a user awakened from a light sleep event because users awakened from a deep sleep event (e.g., REM cycles) are less alert and therefore more likely to fall than users awakened from a light sleep event (e.g., non-REM cycles). As another example, the fall detection threshold can be lowered more for a user awakened from a sleep event that lasts for several hours compared to a user awakened from a sleep event that lasts for a shorter duration (e.g., a nap) because users awakened from longer sleep events are more likely to fall than users awakened from shorter sleep events. As yet another example, the fall detection threshold can be lowered for a user who walks immediately after waking compared to a user who waits a threshold amount of time after waking from a sleep event before walking.

[0030] In some implementations, the fall risk activity can include walking when fatigued. For example, one or more sensors (e.g., accelerometer, gyroscope, etc.) can obtain motion data indicating that the user is engaged in an athletic event (e.g., running, lifting, swimming, cycling, etc.). The risk of the user falling while walking after engaging in the athletic event can vary based on various parameters. For example, the type of athletic event the user engages in can affect the likelihood that the user will fall while walking after completing the athletic event. As an example, the fall risk of a user walking after completing a running workout can be different from the fall risk of a user after completing a swimming workout.

[0031] In some implementations, the fall detection threshold may be adjusted (e.g., lowered) at least in part as a function of the duration of the exercise event. For example, the fall detection threshold may be lowered more for a user after completing a first exercise event of a first duration than for a user after completing a second exercise event of a second duration that is shorter than the first duration.

[0032] In some implementations, an adjusted fall detection threshold for detecting a fall event while a user is engaged in a fall risk activity may be determined using a machine-learned model (e.g., a convolutional neural network). For example, data from one or more sensors (e.g., an accelerometer, a gyroscope, a barometer) may be provided as input to the machine-learned model. The machine-learned model may process the data from the one or more sensors to determine the adjusted fall detection threshold. Further, in some implementations, the adjusted fall detection threshold may be provided as input to a classifier model configured to classify the adjusted fall detection threshold as one of a plurality of fall risk activities. In this manner, the one or more machine-learned models may learn different fall risk activities and different adjusted fall detection thresholds associated with each of the different fall risk activities.

[0033] Wearable computing devices, systems, and methods according to exemplary aspects of the present disclosure may provide numerous technical effects and advantages. For example, the wearable computing device may detect that a user is engaged in a fall risk activity and may adjust (e.g., lower) a fall detection threshold while the user is engaged in the fall risk activity. In this manner, the wearable computing device may be more likely to detect a fall event due to the user being engaged in a fall risk activity. Furthermore, the wearable computing device may provide a notification that the user is at risk for a fall event because the user is at least partially engaged in a fall risk activity. In this manner, the wearable computing device may inform the user about his or her vulnerability to a fall event while engaged in a fall risk activity.

[0034] 1 and 2 are provided illustrating a system 100 for assessing a fall risk of a user 102 in accordance with some embodiments of the present disclosure. The system 100 may include one or more wearable computing devices. For example, in some embodiments, the system may include a first wearable computing device 110 worn on an upper extremity (e.g., wrist) of the user 102, a second wearable computing device 112 worn on a waist of the user 102, and a third wearable computing device 114 worn on a lower extremity (e.g., ankle) of the user 102.

[0035] It should be understood that the system 100 may include more or fewer wearable computing devices than those shown in FIG. 1 . For example, in some implementations, the system 100 may include only one wearable computing device, such as the second wearable computing device 112, that is worn on the wrist of the user 102. It should also be understood that the system 100 may include wearable computing devices that may be worn in locations different than those shown in FIG. 1 . For example, in some implementations, the system 100 may include a wearable computing device that may be worn on the ear of the user 102 (e.g., earphones). Alternatively, or in addition, the system 100 may include a wearable computing device worn on the head of the user 102. For example, in some implementations, the wearable computing device worn on the head of the user 102 may include glasses. In some implementations, the system 100 may include a wearable computing device worn on the foot of the user 102.

[0036] In some implementations, the wearable computing devices 110, 112, 114 may be communicatively coupled to one another via the communications network 200. In this manner, the wearable computing devices 110, 112, 114 may communicate with one another via the communications network 200. Additionally, in some implementations, one or more of the wearable computing devices 110, 112, 114 may be communicatively coupled to a mobile computing device 210 via the communications network 200. Examples of the mobile computing device 210 may include, for example, a smartphone, a tablet, or a laptop. It should be understood that the mobile computing device 210 may include any handheld electronic device having a display screen. In some implementations, the wearable computing devices 110, 112, 114 may be configured to communicate data to a server (not shown) via the communications network 200. For example, in some implementations, the wearable computing devices 110, 112, 114 may be configured to communicate data to a server via the mobile computing device 210. In other words, the wearable computing devices 110, 112, 114 can communicate data to the mobile computing device 210 via the communications network 200, and the mobile computing device 210 can communicate data to a server via the communications network 200 or a different communications network (e.g., the Internet).

[0037] It should also be appreciated that communications network 200 can be any type of communications network, such as a local area network, etc. In general, communications over communications network 200 can occur over any type of wireless connection, using a wide variety of communications protocols, encodings or formats and / or protection schemes (e.g., VPN, secure HTTP, SSL).

[0038] 3, a block diagram of components of a wearable computing device 300 is provided in accordance with some embodiments of the present disclosure. It should be understood that the wearable computing device 300 may be implemented within the system 100 described above with reference to Figures 1 and 2. For example, the wearable computing device 300 may be one of the wearable computing devices 110, 112, 114 worn by the user 102.

[0039] As shown, the wearable computing device 300 may include one or more processors 302. The one or more processors 302 may include any suitable processing device (e.g., a processor core, a microprocessor, an application specific integrated circuit (AISC), a field programmable gate array (FPGA), a microcontroller, etc.). The wearable computing device 300 may further include a memory 304. The memory 304 may include one or more non-transitory computer-readable storage media, such as a random access memory (RAM), a read only memory (ROM), an electronically erasable programmable read only memory (EEPROM), an erasable programmable read only memory (EPROM), a flash memory device, and combinations thereof. The memory 304 may store data 306 and instructions 308, which, when executed by the one or more processors 302, cause the one or more processors 302 to perform the operations disclosed herein.

[0040] The wearable computing device 300 may include a number of sensors 310. For example, in some implementations, the number of sensors 310 may include an accelerometer 312 (e.g., a multi-axis accelerometer) and a gyroscope 314. In this manner, the accelerometer 312, the gyroscope 314, or both, may obtain motion data (e.g., acceleration, angular velocity) indicative of the movement of the user 102 (FIG. 1). For example, in some implementations, the motion data may indicate that the user is engaged in a fall risk activity (e.g., carrying an object, walking after waking up from a nap, etc.). Furthermore, in some implementations, the number of sensors 310 may further include a barometer 316. The barometer 316 may obtain pressure data indicative of a change in altitude. For example, the barometer 316 may obtain pressure data indicative of a change in altitude associated with a user falling (e.g., a fall event).

[0041] In some implementations, the plurality of sensors 310 may include one or more biometric sensors. For example, in some implementations, the one or more biometric sensors may include a photoplethysmogram (PPG) sensor. Alternatively, or in addition, the sensor 310 may include a temperature sensor configured to detect a body temperature of a user 102 ( FIG. 1 ) wearing the wearable computing device 300. It should be understood that the scope of the present disclosure is not intended to cover wearable computing devices with biometric sensors other than a PPG sensor and a body temperature sensor.

[0042] In some implementations, the wearable computing device 300 can include one or more output devices 318. For example, the one or more output devices 318 can include a display screen. In this manner, the wearable computing device 300 can display content (e.g., notifications) viewable by the user 102. Alternatively, or in addition, the one or more output devices 318 can include one or more speakers. In this manner, the wearable computing device 300 can emit audible noises (e.g., alarms, automated voice messages, etc.) for the user 102. As discussed below, the wearable computing device 300 can be configured to assess a fall risk for the user 102.

[0043] In some implementations, the one or more processors 302 may be communicatively coupled to the multiple sensors 310. For example, the one or more processors 302 may be communicatively coupled to the multiple sensors 310 via a data interface (e.g., a data bus). In this manner, the one or more processors 302 may obtain data from the multiple sensors 310. In some implementations, the one or more processors 302 may determine a fall risk activity in which the user 102 (FIG. 1) is engaged based at least in part on the data obtained from one or more of the sensors 310.

[0044] In some implementations, the one or more processors 302 of the wearable computing device 300 may adjust (e.g., lower) a fall detection threshold for detecting whether a fall event (e.g., a fall of a user of the wearable computing device) has occurred. More specifically, the one or more processors 302 may adjust a fall detection threshold of one or more sensors 310 of the wearable computing device 300 based on a fall risk activity in which the user is engaged.

[0045] It should be appreciated that the one or more processors 302 may be configured to adjust the fall detection threshold differently depending on the fall risk activity the user 102 is engaged in. For example, the one or more processors 302 may be configured to lower the fall detection threshold by a first amount when the user is engaged in a first fall risk activity (e.g., walking after waking from a sleep event), and may be further configured to lower the fall detection threshold by a second amount, different from the first amount, when the user is engaged in a second fall risk activity (e.g., walking while carrying an object).

[0046] In some implementations, biometric data obtained from one or more biometric sensors (e.g., PPG sensor, temperature sensor) may indicate that the user 102 is engaged in a fall risk activity. For example, in some implementations, PPG data from the PPG sensor, temperature data obtained from the temperature, or both, may indicate that the user is engaged in a fall risk activity, such as running. In such implementations, a fall detection threshold of one or more sensors 310 (e.g., accelerometer, gyroscope, barometer, etc.) of the wearable computing device 300 may be adjusted (e.g., lowered) in response to data from one or more biometric sensors of the wearable computing device 300 indicating that the user is engaged in a fall risk activity, such as running.

[0047] In some implementations, biometric data obtained from one or more biometric sensors (e.g., PPG sensor, temperature sensor) of the wearable computing device 300 may indicate that the user 102 is not engaged in a low-risk activity (e.g., sitting). For example, in some implementations, the biometric data may indicate that the user 102 is sitting. In such implementations, the one or more processors 302 of the wearable computing device 300 may be configured to adjust (e.g., increase) a fall detection threshold of the wearable computing device 300 because the user 102 is engaged in a low-risk activity.

[0048] In some implementations, biometric data obtained from one or more biometric sensors of the wearable computing device 300 may indicate whether the wearable computing device 300 is currently being worn by the user 102. For example, in some implementations, biometric data obtained from the one or more biometric sensors may indicate that the wearable computing device 300 is not being worn by the user 102 (e.g., off the wrist). In such implementations, the one or more processors 302 may be configured to disable the fall detection functionality while the biometric data indicates that the wearable computing device 300 is not being worn by the user 102. In this manner, erroneous data from the one or more sensors 310 indicating that the user 102 is engaged in a fall risk activity may be ignored. It should be appreciated that the one or more computing devices 302 may be configured to enable the fall detection functionality when the biometric data indicates that the wearable computing device 300 is being worn by the user 102 (e.g., on the wrist).

[0049] In some implementations, the wearable computing device 300 may include one or more machine-learned models 320. For example, in some implementations, the one or more machine-learned models 320 may be stored in the memory 304 of the wearable computing device 300. In alternative implementations, the one or more machine-learned models 320 may be stored in the memory of one or more devices that are remote to the wearable computing device 300. For example, in some implementations, the one or more machine-learned models 320 may be stored in the memory of a mobile computing device 210 (FIG. 2) communicatively coupled to the wearable computing device 300 via the network 200. Alternatively, or in addition, the one or more machine-learned models 320 may be stored in one or more servers (not shown) communicatively coupled to the wearable computing device 300 via the network 200. As described below, the one or more machine-learned models 320 may be configured to adjust a fall detection threshold based at least in part on data obtained from one or more of the sensors 310.

[0050] 4, data from one or more of the sensors 310 may be provided as input to one or more machine-learned models 320. For example, in some implementations, data from at least one of the accelerometer 312 (FIG. 3), gyroscope 314 (FIG. 3), or barometer 316 may be provided as input to the one or more machine-learned models 320. The one or more machine-learned models 320 may process data from the one or more sensors 310 to determine an adjusted fall detection threshold for a fall event. In some implementations, an output of the one or more machine-learned models 320 may be a single numerical value that may be used to determine whether a fall event has occurred. In alternative implementations, the output of the one or more machine-learned models 320 may include multiple outputs. Each of the multiple outputs may correspond to a fall detection threshold of one or more of the sensors 310. For example, a first output of the one or more machine-learned models 320 may correspond to a first fall detection threshold of the accelerometer 312. Conversely, a second output of the one or more machine learned models 320 may correspond to a second fall detection threshold of the gyroscope 314. Still further, in some implementations, a third output of the one or more machine learned models 320 may correspond to a third fall detection threshold of the barometer 316.

[0051] It should be appreciated that the one or more machine learned models 320 can include any suitable type of machine learned model. For example, the one or more machine learned models 320 can include, without limitation, a convolutional neural network, a decision tree, a Bayesian network, a support vector machine, a K-means cluster, or any other suitable type of machine learned model.

[0052]

[0031] Referring now to Figure 5, a flow diagram of an exemplary method 500 for assessing a risk of a fall for a user wearing a wearable computing device is provided, according to some embodiments of the present disclosure. Method 500 may be implemented, for example, using the wearable computing device 300 described above with reference to Figure 3. Figure 5 shows steps performed in a particular order for purposes of illustration and explanation. Those skilled in the art, using the disclosure provided herein, will appreciate that the various steps of method 500, or any of the other methods disclosed herein, may be adapted, modified, rearranged, performed simultaneously, or modified in various ways without departing from the scope of the present disclosure.

[0053] At (502), method 500 may include acquiring, by one or more processors of a wearable computing device worn by the user, data indicating that the user is engaged in a fall risk activity. In some implementations, the data may be acquired from a mobile computing device (e.g., a smartphone, a tablet, etc.) associated with the user. For example, the data may include a signal indicating that the user is engaged in a fall risk activity (e.g., walking while looking at a display of the mobile computing device). In this manner, the mobile computing device may determine that the user is engaged in a fall risk activity and may communicate the data (e.g., a signal) indicating that the user is engaged in the fall risk activity to the wearable computing device.

[0054] In alternative implementations, the data may be obtained from one or more sensors of the wearable computing device. For example, an accelerometer of the wearable computing device may obtain acceleration data along one or more axes that indicate the user is engaged in a fall risk activity. For example, the acceleration data may indicate that the user is walking without moving the upper limbs, which may indicate that the user is holding an object (e.g., a smartphone, tablet, box, etc.) while walking. Alternatively, or additionally, in some implementations, the data may be obtained from one or more sensors of another wearable device worn by the user. For example, the data may be obtained from one or more sensors of a wearable computing device worn, for example, on the user's lower limbs (e.g., ankles).

[0055] At (504), method 500 may include adjusting a fall detection threshold of the wearable computing device based at least in part on the data obtained at (502). For example, in some implementations, the fall detection threshold may be lowered. In this manner, the wearable computing device may be more sensitive to detecting a fall event while the user is performing a fall risk activity.

[0056] In some implementations, adjusting the fall detection threshold of the wearable computing device may be a two-step process. For example, in some implementations, adjusting the fall detection threshold of the wearable computing device may include providing the data acquired at (502) as an input to a first machine-learned model configured to classify the fall risk activity as one of a plurality of different fall risk activities (e.g., carrying a box, walking after waking from sleep, etc.). In other words, the machine-learned model may be configured to output a classification of the fall risk activity. For example, in some implementations, the machine-learned model may be configured to output an identifier of one of a plurality of different fall risk activities into which the machine-learned model classifies the data acquired at (502).

[0057] In some implementations, the classification of the fall risk activity may be provided as an input to a second machine-learned model. The second machine-learned model may be configured to process the input to determine an adjusted fall detection threshold, which is provided as an output of the second machine-learned model. For example, in some implementations, the output may be a single adjusted fall detection threshold. In alternative implementations, the output may be a plurality of adjusted fall detection thresholds. Furthermore, in such implementations, each of the plurality of adjusted fall detection thresholds may be associated with a different sensor. For example, a first adjusted fall detection threshold may be associated with a first type of sensor (e.g., accelerometer) of the wearable computing device, while a second adjusted fall detection threshold may be associated with a second type of sensor (e.g., gyroscope, barometer) of the wearable computing device.

[0058] At (506), method 500 may include providing a notification indicating that the user is at risk of falling because the user is at least partially engaged in a fall risk activity. For example, in some implementations, the wearable computing device may include an output device (e.g., a piezoelectric motor) that may vibrate to prompt the user to put down the object and read a notification provided on a display screen of the wearable computing device indicating the fall risk activity. In alternative implementations, the wearable computing device may emit an audible notification (e.g., a computer-generated voice) indicating the fall risk activity.

[0059]

[0036] Referring now to Figure 6, a flow diagram of an example method 600 for adjusting a fall detection threshold of a wearable computing device is provided, according to some embodiments of the present disclosure. Method 600 may be implemented, for example, using the wearable computing device 300 described above with reference to Figure 3. Figure 6 shows steps performed in a particular order for purposes of illustration and explanation. Those skilled in the art, using the disclosure provided herein, will appreciate that the various steps of method 600, or any of the other methods disclosed herein, may be adapted, modified, rearranged, performed simultaneously, or modified in various ways without departing from the scope of the present disclosure.

[0060] At (602), the method 600 may include providing data from one or more sensors of one or more wearable computing devices worn by a user, where the data may be provided as an input to the machine-learned model. For example, in some implementations, acceleration data over a period of time may be provided as a first input to the machine-learned model. Additionally, in some implementations, pressure data from a barometer of one or more wearable computing devices may be provided as a second input to the machine-learned model. It should be appreciated that data from any suitable sensor of one or more computing devices capable of obtaining data indicative of a user's movements may be provided as an input to the machine-learned model.

[0061] At (604), method 600 may include processing the data input at (602) to determine an adjusted fall detection threshold. For example, in some implementations, the adjusted fall detection threshold can be a single numerical value that can be used to determine whether a user's movement corresponds to a fall event. At (606), method 600 may include outputting the adjusted fall detection threshold.

[0062]

[0031] Referring now to Figure 7, a flow diagram of another method for assessing a risk of a fall of a user wearing a wearable computing device according to some embodiments of the present disclosure is provided. Method 700 may be implemented using, for example, the wearable computing device 300 described above with reference to Figure 3. Figure 7 shows steps performed in a particular order for purposes of illustration and explanation. Those skilled in the art will appreciate, using the disclosure provided herein, that the various steps of method 700, or any of the other methods disclosed herein, may be adapted, modified, rearranged, performed simultaneously, or modified in various ways without departing from the scope of the present disclosure.

[0063] At (702), the method 700 may include obtaining sleep data for a sleep event associated with a user wearing the wearable computing device. For example, the sleep data may include a duration of the sleep event. Alternatively, or in addition, the sleep data may indicate a depth of the sleep event. For example, the depth of the sleep event may indicate whether the user is waking from a deep sleep event (e.g., a REM sleep cycle) or a light sleep event (e.g., a non-REM sleep cycle). It should be appreciated that the sleep data may include any data indicative of one or more metrics associated with a sleep event in which the user is awake.

[0064] At (704), the method 700 can obtain data indicating that the user is engaging in a fall risk activity after waking from the sleep event. For example, the fall risk activity can include walking after waking from the sleep event. In some implementations, the data can include data from one or more sensors of the wearable computing device indicating that the user is walking. For example, the data can include data indicating the pace at which the user is walking.

[0065] At (706), the method 700 may include adjusting a fall detection threshold based on the sleep data acquired at (702) and the data acquired at (704) indicating that the user is engaged in a fall risk activity. It should be appreciated that the fall detection threshold may be adjusted differently depending on the data acquired at (702) and (704). For example, depending on the duration of the sleep event, the fall detection threshold may be adjusted differently (e.g., lowered). In some implementations, the fall detection threshold may be lower for a user awakened from a sleep event lasting several hours compared to a user awakened from a shorter sleep event, such as a nap. In this manner, the wearable computing device may be more sensitive to detecting a fall event for a user awakened from a sleep event lasting several hours compared to a user awakened from a sleep event of shorter duration.

[0066] In some implementations, the fall detection threshold may be adjusted (e.g., lowered) differently depending on the depth of the sleep event. For example, the fall detection threshold may be lower for a user awakened from a deep sleep event (e.g., a REM cycle) than for a user awakened from a lighter sleep event (e.g., a non-REM cycle). In this way, the wearable computing device may be more sensitive to detecting a fall event for a user awakened from a deeper sleep event compared to a user awakened from a lighter sleep event.

[0067] In some implementations, the fall detection threshold may be further lowered for a user walking at a pace faster than a threshold pace associated with walking after awakening from a sleep event. In this manner, the wearable computing device may be more sensitive to detecting a fall event when the user is walking at a pace above the threshold pace.

[0068] At (708), the method 700 can include providing a notification indicating that the user is at risk of falling while walking after awakening from a sleep event. For example, in some implementations, the notification can include a visual notification provided via a display screen of a wearable computing device worn by the user. Alternatively, or additionally, the notification can include an auditory notification provided via a speaker of a wearable computing device worn by the user. In this manner, the notification (e.g., visual, auditory, or both) can prompt the user to walk more cautiously (e.g., at a pace below a threshold pace associated with walking after awakening from a sleep event).

[0069] 8 and 9, a wearable computing device 800 is provided in accordance with some embodiments of the present disclosure. As shown, the wearable computing device 800 may be worn, for example, on a user's arm (e.g., wrist). For example, the wearable computing device 800 may include a housing 810. The wearable computing device 800 may include one or more electronic components (e.g., disposed on a printed circuit board) disposed within the housing 810. Additionally, the wearable computing device 800 may include a battery (not shown) disposed within the housing 810.

[0070] As shown, the wearable computing device 800 can include a first band 820 coupled to the housing 810 at a first location and a second band 822 coupled to the housing 810 at a second location. The first band 820 and the second band 822 can be coupled to one another to secure the housing 810 to a user's arm. For example, the first band 820 can include a buckle or clasp (not shown). Additionally, the second band 822 can define a plurality of apertures 824 spaced apart from one another along a length of the second band 822. In such an implementation, a protrusion of a buckle associated with the first band 820 can extend through one of the plurality of apertures defined by the second band 822 to couple the first band 820 to the second band 822.

[0071] It should be appreciated that any suitable type of fastener may be used to couple the first band 820 to the second band 822. For example, in some implementations, the first band 820 and the second band 822 may include magnets. In such implementations, the first band 820 and the second band 822 may be magnetically coupled to one another to secure the housing 810 to the user's arm.

[0072] In some implementations, the wearable computing device 800 may include a display 830 configured to display content for a user to view (e.g., time, date, biometrics, notifications, etc.). For example, the display 830 may include a number of pixels. In some implementations, the display 830 may include an organic light emitting diode (OLED) display. However, it should be understood that the display 830 may include any suitable type of display.

[0073] In some implementations, the wearable computing device 800 may include a first electrode 840 and a second electrode 842. It should be understood that the wearable computing device 800 may include more or fewer electrodes. As shown, the first electrode 840 and the second electrode 842 may be positioned with respective openings (e.g., cutouts) defined by the housing 810. Additionally, the first electrode 840 and the second electrode 842 may each contact (e.g., touch) the user's wrist. In this manner, the first electrode 840 and the second electrode 842 may be used to measure one or more biometrics (e.g., electrodermal activity, electrocardiogram) of the user.

[0074] 10, the wearable computing devices 110, 112, 114 may be communicatively coupled to a network 900. The wearable computing devices 110, 112, 114 may communicate data (e.g., motion data) via the network 900 to a remote computing system 910 (e.g., an Internet of Things device) for storage and / or processing. For example, in some embodiments, the wearable computing devices 110, 112, 114 may communicate data to the mobile computing device 210. In such embodiments, the wearable computing devices 110, 112, 114 may communicate data to the mobile computing device 210, which may then communicate data via the network 900 to the remote computing system 910. In alternative embodiments, the wearable computing devices 110, 112, 114 may bypass the mobile computing device 210 and instead communicate data directly to the remote computing system 910 via the network 900.

[0075] The remote computing system 910 includes one or more processors 912 and memory 914. The one or more processors 912 can be any suitable processing device (e.g., a processor core, a microprocessor, an ASIC, an FPGA, a controller, a microcontroller, etc.) and can be a single processor or multiple processors operably connected. The memory 914 can include one or more non-transitory computer-readable storage media, such as RAM, ROM, EEPROM, EPROM, flash memory devices, magnetic disks, etc., and combinations thereof. The memory 914 can store data 916 and instructions 918 that are executed by the processor 912 to cause the remote computing system 910 to perform operations, such as any of the operations described herein. For example, in some implementations, the memory 914 of the remote computing system 910 can be configured to store one or more machine-learned models 320 described above with reference to FIG. 3. In this manner, data obtained from one or more sensors (e.g., accelerometer, gyroscope, barometer, etc.) on board the wearable computing device 110, 112, 114 and indicative of a user engaging in a fall risk activity can be communicated to the remote computing system 910 and provided as input to one or more machine-learned models 320 stored in memory 914 thereof. The one or more machine-learned models 320 can be configured to process the data and output an adjusted fall detection threshold. Further, the adjusted fall detection threshold can be communicated to the wearable computing device 110, 112, 114 via the network 900. In this manner, the fall detection threshold can be adjusted while a user wearing the wearable computing device 110, 112, 114 is engaged in a fall risk activity.

[0076] In some embodiments, the remote computing system 910 includes or is otherwise implemented by one or more computing devices. If the remote computing system 910 includes multiple server computing devices, such server computing devices may operate according to a sequential computing architecture, a parallel computing architecture, or some combination thereof.

[0077] Network 900 may be any type of communications network, such as a local area network (e.g., an intranet), a wide area network (e.g., the Internet), or some combination thereof, and may include any number of wired or wireless links. In general, communications over network 900 may occur over any type of wired or wireless links, using a wide variety of communications protocols (e.g., TCP / IP, HTTP, SMTP, FTP), encodings or formats (e.g., HTML, XML), and / or protection schemes (e.g., VPN, Secure HTTP, SSL).

[0078] The technology described herein refers to sensors and other computer-based systems, and the actions performed and information transmitted to and from such systems. Those skilled in the art will recognize that the inherent flexibility of computer-based systems allows for a wide variety of possible configurations, combinations, and divisions of tasks and functions between and among components. For example, the server processes described herein may be implemented using a single server or multiple servers working in combination. Databases and applications may be implemented on a single system or distributed across multiple systems. Distributed components may operate serially or in parallel.

[0079] Although the subject matter of the present disclosure has been described in detail with respect to various specific and exemplary embodiments thereof, each example is provided for the purpose of illustration and not for the purpose of limiting the present disclosure. Those skilled in the art, upon understanding the above content, can easily make modifications, variations, and equivalents of such embodiments. Thus, the present disclosure does not exclude the inclusion of such modifications, variations, and / or additions to the subject matter of the present disclosure as would be readily apparent to those skilled in the art. For example, features illustrated or described as part of one embodiment can be used with another embodiment to produce yet another embodiment. Thus, the present disclosure is intended to cover such modifications, variations, and equivalents.

Claims

1. 1. A computer-implemented method for assessing a user's risk of falling, comprising: obtaining, by one or more processors, data indicative of the user engaging in a fall risk activity; adjusting, by the one or more processors, a fall detection threshold of a wearable computing device worn by the user based at least in part on the data indicating that the user is engaged in the fall risk activity; and providing, by the one or more processors, a notification indicating that the user is at risk of falling due, at least in part, to the user engaging in the fall risk activity.

2. 2. The computer-implemented method of claim 1, wherein obtaining data indicative of the user engaging in the fall risk activity includes obtaining a signal by the one or more processors from a mobile computing device associated with the user, the signal indicating the user engaging in the fall risk activity.

3. The computer-implemented method of claim 2 , wherein providing the notification comprises providing, by the one or more processors, the notification for display on a display screen of the mobile computing device.

4. 2. The computer-implemented method of claim 1, wherein obtaining the data indicative of the user engaging in the fall risk activity comprises obtaining data from one or more sensors of the wearable computing device worn by the user.

5. Adjusting the fall detection threshold of the wearable computing device includes: providing, by the one or more processors, the data from the one or more sensors of the wearable computing device as input to a machine-learned model configured to process the data from the one or more sensors of the wearable computing device to determine an adjusted fall detection threshold; and obtaining, by the one or more processors, the adjusted fall detection threshold as an output of the machine-learned model.

6. the one or more sensors of the wearable computing device include a plurality of different sensors; 6. The computer-implemented method of claim 5, wherein the adjusted fall detection threshold comprises a plurality of adjusted fall detection thresholds, each of the plurality of adjusted fall detection thresholds associated with a respective sensor of the plurality of different sensors.

7. the fall risk activity comprises walking after waking from a sleep event; 2. The computer-implemented method of claim 1, wherein the data indicative of the user engaging in the fall risk activity comprises movement data indicative of the user walking rapidly after awakening from the sleep event at a pace above a threshold pace.

8. The computer-implemented method of claim 7 , further comprising obtaining, by the one or more processors, sleep data associated with the sleep event.

9. 10. The computer-implemented method of claim 8, wherein adjusting the fall detection threshold of the wearable computing device includes adjusting, by the one or more processors, the fall detection threshold of the wearable computing device according to the sleep data and the movement data.

10. 10. The computer-implemented method of claim 9, wherein the sleep data includes at least one of a duration of the sleep event or a depth of the sleep event.

11. 11. The computer-implemented method of claim 10, wherein the depth of the sleep event includes data indicative of whether the user is arousing from a rapid eye movement (REM) sleep cycle or arousing from a non-REM sleep cycle.

12. Obtaining data indicative of the user engaging in a fall risk activity includes: acquiring, by the one or more processors, first data from one or more sensors of a first wearable computing device worn at a first location on the user's body; and acquiring, by the one or more processors, second data from one or more sensors of a second wearable computing device worn at a second location on the body of the user that is different from the first location.

13. 13. The computer-implemented method of claim 12, wherein adjusting the fall detection threshold of the wearable computing device includes adjusting, by the one or more processors, the fall detection threshold of at least one of the first wearable computing device or the second wearable computing device.

14. 1. A wearable computing device, comprising: A plurality of sensors; one or more processors communicatively coupled to the plurality of sensors; The one or more processors: acquiring data indicative of a user wearing the wearable computing device engaging in a fall risk activity; adjusting a fall detection threshold based at least in part on the data indicating that the user is engaged in the fall risk activity; The wearable computing device is configured to provide a notification indicating that the user is at risk of falling due, at least in part, to the user engaging in the fall risk activity.

15. To obtain data indicative of the user engaging in the fall risk activity, the one or more processors: providing data from one or more sensors of the plurality of sensors as input to a machine-learned model configured to classify the data from the one or more sensors as corresponding to a first fall risk activity of a plurality of different fall risk activities; The wearable computing device of claim 14 , configured to obtain an identifier indicative of the first fall risk activity as an output of the machine-learned model.

16. the data indicative of the user engaging in the fall risk activity is obtained from one or more sensors of the plurality of sensors; To adjust the fall detection threshold, the one or more processors: providing the data from the one or more sensors of the plurality of sensors as input to a machine-learned model; The wearable computing device of claim 14 , configured to obtain an adjusted fall detection threshold as an output of the machine-learned model.

17. The wearable computing device of claim 14 , wherein the plurality of sensors includes at least one of an accelerometer, a gyroscope, or a barometer.

18. The wearable computing device of claim 14 , further comprising one or more output devices configured to output the notification indicating that the user is at risk of falling.

19. 20. The wearable computing device of claim 18, wherein the one or more output devices comprise at least one of a speaker or a display screen.

20. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations, the operations including: obtaining data indicative of a user engaging in a fall risk activity; adjusting a fall detection threshold based at least in part on the data indicating that the user is engaged in the fall risk activity; and and providing a notification indicating that the user is at risk of falling due, at least in part, to the user's engagement in the fall risk activity.

Images