Self-controlled wearable device
By combining non-invasive wearable devices with cloud systems and utilizing intelligent schedulers and multi-communication stacks to manage data transmission, the problem of high battery consumption in wearable devices has been solved. This enables efficient and continuous monitoring of the physiological state of the battery and collaborative work among multiple devices, improving the flexibility and accuracy of data transmission.
Patent Information
- Application Number
- CN202280088404.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2021-12-10
- Filing Date
- 2022-12-12
- Publication Date
- 2024-11-19
AI Technical Summary
Existing wearable devices consume a lot of battery when monitoring biometrics and are not suitable for long-term use. They also mostly rely on external devices for calculation and data exchange, which limits their functionality.
It combines non-invasive wearable devices with a cloud system, manages data transmission through an intelligent scheduler and multiple communication stacks, stores and processes physiological data locally on the device, dynamically selects communication channels to optimize battery usage, and supports collaborative work among multiple devices.
It enables efficient and continuous monitoring of battery physiological status, improves the flexibility and accuracy of data transmission, supports seamless data exchange between multiple devices, and protects the privacy of user information.
Smart Images

Figure CN118984674A_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This patent application claims the benefit of U.S. Provisional Application No. 63 / 288,300, filed on December 10, 2021, which is incorporated herein by reference in its entirety. Technical Field
[0003] The present invention relates to the fields of non-invasive digital health monitoring, physiological signal processing and biological data computing. Background Art
[0004] The health benefits derived from tracking biometrics over time are gaining traction in the consumer wearable space and are also gaining recognition among clinical professionals. Wearable devices are now used daily by consumers who can track their health with one of the many devices available on the market. However, such wearable devices have only limited functionality when they operate independently and rely heavily on external edge computing devices (such as mobile phones) to perform most computing and data exchange tasks. On the other hand, devices that both retrieve and analyze data internally are battery intensive, slow, and unsuitable for continuous health monitoring over long periods of time.
[0005] In clinical settings, patient monitoring is typically invasive, expensive, and often limited to the time a patient is in a hospital room or healthcare setting. Aside from the already commercially available wearable devices that collect biometric data over a very small area, there are few commercially available devices for measuring a person's physiology without the use of a person or invasive equipment.
[0006] Therefore, there is a need for more robust wearable devices that can monitor and collect a wider range of individual biometric data in a non-invasive manner. Summary of the invention
[0007] The claimed invention is directed to providing systems and methods for non-invasive health monitoring in a battery efficient and sustained manner. In one aspect, the system is configured to continuously monitor the physiological state of a user by using a cloud configured to monitor and analyze physiological data related to the user and a non-invasive instrument used as a data acquisition device used by the user. The device may include a power source, at least one sensor, a wireless communication device configured to communicate with the cloud, a memory, and at least one processor. The processor may be configured to determine at least one biological metric from the data collected by the sensor, communicate the at least one biological metric to the cloud, and determine when to send the at least one biological metric to the cloud based on the state of the power source and the characteristics of the at least one biological metric. In addition, the cloud may be configured to process at least one biological metric from the data acquisition device to determine the current health state of the user.
[0008] On the one hand, the non-invasive instrument is capable of near field communication, thereby transferring data from a first device to a second device, and is capable of communicating with a second nearby device. In addition, the system is configured to identify the source of the sensor data, and the non-invasive device is capable of determining the proximity relationship between more than one sensor data relative to each other, and determining the source of sensor data about the user. On the other hand, the non-invasive device stores user profile information including user information, algorithm information, and device information. In these aspects, the system can be configured to retrieve the user profile information and send data through a communication channel, and the system determines the best communication channel for a specific communication based on one or more metrics, including but not limited to the state of the device, the battery power available on the device, the battery power required to complete the communication, the processing power of the device, and the processing power required to complete the communication. The algorithm information can also be configured to allow the device to filter out the data to be sent and collected (aggregated and compressed), thereby further reducing battery consumption and ensuring continuity. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] Figure 1 It is a wearable device according to the prior art.
[0010] Figure 2 is an overview of using a wearable device in a clinical setting according to an aspect of the present invention.
[0011] Figure 3 A wearable device according to one aspect of the present invention.
[0012] Figure 4 is a schematic representation of different triggers for data to be sent.
[0013] Figure 5It is a schematic representation of transferring user information from one wearable device to another. DETAILED DESCRIPTION
[0014] Embodiments of the present invention will be described more fully below with reference to the accompanying drawings showing embodiments of the present invention. However, the present invention can be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein; on the contrary, these embodiments are provided so that the present disclosure will be comprehensive and complete and will fully convey the scope of the present invention to those skilled in the art. Various changes, modifications, and equivalents of the methods, devices, and / or systems described herein will be apparent to those of ordinary skill in the art. The sequences described herein are merely examples, are not limited to the embodiments set forth herein, and may be changed, as will be apparent to those of ordinary skill in the art.
[0015] In the following description, many specific details are set forth. However, it will be appreciated that embodiments of the present invention can be practiced without these specific details. In order to avoid obscuring the understanding of the specification, in other examples, well-known methods, structures, and techniques have been shown in detail. However, descriptions of functions and structures well known to those of ordinary skill in the art may be omitted to improve clarity and brevity.
[0016] definition
[0017] Core - Server, where wearable data streams as well as biometrics are aggregated and processed into health insights.
[0018] Central site (consulate) - Standardized interface / software that can be integrated on the cloud of any company using wearable devices and that enables two-way communication between wearable devices and servers.
[0019] Communication stack - also known as protocol stack. A stack is a software implementation of a set of layered communication protocols that operate together to provide a set of network functions.
[0020] On the one hand, the wearable device is configured to save the real-time status of the user's biometrics, the user's physiological state (e.g., sedentary, light activity, exercise, sleeping, etc.), and events that may require the immediate attention of another person (e.g., falls, out-of-range biometrics such as heart rate or breathing rate). These states and data are transmitted via the communication means available to the wearable device. In addition, the information generated by the device (from data sets of long time periods with low urgency about sleep periods to short events of high urgency such as possible falls) is assigned additional metadata (such as the urgency of the data and the maximum validity period). However, each communication channel / option of the wearable device (e.g., Bluetooth, WiFi, cellular network) has an associated specific cost during their use. These costs include: (a) how much power is consumed when using the channel, which directly affects the battery consumption, (b) the throughput of the channel (e.g., how many bytes of information per second), and (c) the actual monetary cost (e.g., the cost of the cellular network data plan).
[0021] The wearable device uses state information, communication channel power consumption and cost, and data urgency and validity period to evaluate how quickly the device's information needs to be transmitted and which appropriate communication channel to use. For example, the device detects a possible fall. The data required to classify the fall is moderate and the urgency is high. The device will utilize any available communication channel that guarantees immediate transmission, regardless of the impact of power consumption or data loss. In another embodiment, the wearable device completes the filling of a fixed-size file that contains standard longitudinal biometrics that will be aggregated and summarized every 24 hours. The data has a long validity period and low urgency. Even if no other options are available, the device will avoid using higher-cost communication channels (such as cellular networks) until another option such as Bluetooth or WiFi becomes available. These and other aspects are disclosed below.
[0022] One embodiment of the claimed invention includes a system 10 for continuously monitoring a user's physiological state, the system comprising a significantly improved wearable device 105, such as Figure 3As shown. The wearable device 105 is capable of controlling data exchange with the cloud (e.g., core 103 / application backend 102) and the internal behavior of the wearable device 105 itself. In one aspect, the wearable device 105 includes a power source (e.g., a rechargeable battery), a sensor (e.g., PPG, accelerometer, etc.), a memory, a wireless communication device (e.g., Wi-Fi, Bluetooth, NFC, etc.), and a processor. The wearable device 105 can be configured to determine the user's biometrics from the data collected by the sensor. In one aspect, the wearable device 105 can be configured to capture and obtain physiological data from the user in a manner similar to those disclosed in U.S. Patent No. 9,820,656 (issued on November 21, 2017), U.S. Patent No. 11,129,568 (issued on September 28, 2021), and U.S. Patent No. 11,291,392 (issued on April 5, 2022), which are incorporated herein by reference in their entirety.
[0023] The wearable device 105 utilizes a configurable scheduler on the device to transfer data between the device 105 and the cloud 102 / 103. The wearable device 105 has real-time access to the user's physiological data and can ensure that the data is analyzed and sent to the right location while the data remains relevant by utilizing a set of pre-programmed rules downloaded from the cloud to the device or learned by the device 105. Decisions made on the wearable device 105 can be determined by pre-programmed or device-learned rules.
[0024] In one aspect, the scheduler is as simple as the client, the scheduler needs a web portal dashboard that updates the biometric information every 15 minutes, unless when the person's biometric deviates outside of the standard range for that person, in which case new measurements are wanted every second (1Hz). The scheduler will be configured via instructions from the cloud, with a portal telling the scheduler to transmit all data collected on the device every 15 minutes (roughly a quarter of an hour), monitor the biometric against the standard range for the specific user, and immediately transmit 1Hz data when the biometric is outside of the standard range. The scheduler can and will have multiple portals for multiple data types and specific events or states, all of which can be remotely updated / deleted / added to the device from the cloud to change the behavior of the scheduler.
[0025] In the prior art system, the external device or other application determines when to call the wearable device to obtain data (see Figure 1). According to one aspect of the invention, the wearable device 105 is configured to support a range of communication stacks and determines which communication stack to use at a given time by utilizing a set of pre-programmed rules downloaded to the device from the cloud or learned by the device. Thus, instead of waking up an application on the external device 106 or relying on another application to drive the data flow, the wearable device 105 can determine which data needs to be sent and how it will be communicated (see Figure 4 ).
[0026] The wearable device 105 is configured to store the user state (i.e., physiological state, such as sedentary, light movement, exercise, sleeping, etc.) in internal memory and in the cloud (core 103 / application backend 102), which allows for transmission between devices (see Figure 5 ), while charging the battery or while performing software updates, continuously collecting data. This is done by creating a user profile that is stored locally on the wearable device, further allowing the device to store data until the data can / should be communicated.
[0027] Although the data bridge network 107 ( Figure 3 ) to preserve the confidentiality of personally identifiable information (PII), but the wearable device 105 is able to associate the user with the data source via the user profile (device and user are associated, linked by the user profile), and the data source is easily accessible and transferable, allowing the wearable device 105 to be exchanged between multiple users. In addition, the ability to preserve the confidentiality of PII allows different devices 405 and 406 within the system 10 that are not necessarily from the same company ( Figure 5 ) use and connection. This also allows for the use of more than one wearable device (simultaneously or alternatively) to continuously collect data from a single user. The ability to use multiple devices (including the ability to use multiple devices simultaneously) allows for continuous and distributed sensing. If two devices are simultaneously collecting information about a single user from different locations on the user, the device is able to achieve more functionality in terms of data analysis and health identification / prediction by dynamically selecting sensor data from the best available location. In addition, the device can be configured to communicate with other sensor devices and retrieve data from such sources. Wireless communication and the ability to locate nearby devices allow devices to perform contact tracing between themselves, other nearby wearable devices, and other machines that can communicate with the device and / or triangulate the device's location.
[0028] The ability of the device 105 / 405 / 406 to collect data and run algorithms allows the device to filter out (i.e., aggregate and / or compress the high-resolution data in the device) the data to be sent and collected, further reducing battery consumption and ensuring continuity. By aggregating / compressing the data into a more compact but still relevant data summary, less data will be sent, requiring a smaller and shorter transmission window, requiring less power. By giving the device the ability to determine which analysis to perform and where to perform the analysis, accuracy is also improved. The various unique features of the claimed invention are described in detail below.
[0029] Intelligent communication with applications
[0030] Most wearable devices 205 of prior art systems 20 deliver data via cloud-app push notifications. These notifications wake up the app on the user's smartphone 201, while the backend / cloud 202 / 203 pulls data from the wearable device 205 using Bluetooth or Bluetooth Low Energy (BLE) (see Figure 1 ). The wearable device 105 of the present invention 10 itself (see Figure 3 ) controls data exchange / flow and its behavior without using push notifications. The device 105 runs a configurable scheduler to initiate transmissions from the device 105 to the core 103 / application backend 102 / cloud. This allows for more power-efficient use of the communication stack (such as Bluetooth or BLE), but can include higher power options such as Wi-Fi, and allows the device 105 to take advantage of higher throughput data transmissions. The device 105 uses real-time contextual knowledge and is able to run logic to make those decisions. The device 105 has a routing engine that follows a set of configurable rules based on the connectivity options available to the wireless device 105, and the cost of the connectivity options is as described above (e.g., only transmit data when in a certain geographic area, transmit data via Wi-Fi when Wi-Fi at a certain bandwidth is detected, etc.). This is only done when the logic is executed on the device 105 rather than in the cloud.
[0031] Using the algorithm on the device, the wearable device 105 can create a rule group to override the scheduler and push data from the device 105 to the cloud 102 / 103 when needed. For example, the device 105 can use the scheduler until a health emergency of the user occurs. When the wearable device 105 is configured to monitor the physiological parameters of an individual (e.g., heart rate, respiratory rate, SpO, etc.), the wearable device 105 can determine the normal physiological parameters of the user and send information when the wearable device 105 detects any deviation from those ranges. For example, the device 105 is configured to learn by observing that the normal respiratory rate of a person falls between X and Y. As long as the real-time observation is within this range, the scheduler is used. However, if the respiratory rate is outside the normal range, there is a scheduler override to quickly track the sending data. In other words, the wearable device 105 can be configured to send data according to the instructions of the scheduler until some unexpected event occurs that is bound to the wearer of the wearable device. Other embodiments include fall detectors, blood oxygen detectors, emergency buttons, users entering dangerous geo-fenced areas, etc.
[0032] This method of initiating the delivery from the device 105 is critical because initiating wireless communication is a major factor in battery consumption. Streaming data in real time drains the battery, which is counterintuitive for a product that is intended to run continuously. Therefore, the present invention runs on a scheduler so that it does not drain the battery. By doing so, the system 10 also eliminates the need to utilize push notifications. However, as Figure 4 As shown, the system is flexible: the system is not configured to receive information only when the scheduler is triggered - the scheduler can be overridden to push data when there is an abnormal value or a predetermined outlier that the device is searching for. The system 10 uses real-time contextual knowledge to run processing on the device 105; the device 105 itself can then determine when to send information, that is, to comply with the schedule determined by the scheduler, or to send information when needed (e.g., alarms, etc.). The threshold / outlier can be determined by the device 105 itself. In other aspects, the device 105 may follow preset rules, such as sending data when a threshold is met or an outlier is detected, but it is up to the device 105 to determine what the threshold / outlier is. If the physiological measurement triggers an abnormal value or an outlier, the device will send data outside the normal scheduler. In this way, the device has both regular delivery and "override" delivery. The device can use any physiological measurement value / combination to determine the state that will trigger the override. In addition, the device may utilize non-physiological states, and the device can collect information such as physiological metrics, quality control, device positioning, etc.
[0033] In general, the device 105 has access to a complete range of information in real time, including but not limited to physiological data associated with the wearer of the device and the status of the device itself (e.g., battery status, memory status, storage space, etc.). The system 10 can provide rules to the device 105 and be specified on a per-device basis to inform the device to change the rules for notification and scheduling. For example, the system 10 can be utilized by a mining company that needs specific rules, such as when to override the scheduler when an abnormal value occurs related to oxygen saturation level, or if the battery is below a certain percentage, then do not send data. In a frail care setting, the rules may utilize a fall detector, and when the fall detector detects a fall, then the device should push data so that the monitoring center can analyze and respond to the fall of the wearer of the device. The key here is flexibility: the device 105 can be updated with modified rules that determine how to decide when to push data based on the specific needs and expectations of the user. The reason flexibility is very important is because some data has a certain "expiration date". For example, when someone's heart stops, it is useless to notify this change when the device 105 is connected to an external power source 12 hours later. Therefore, the goal of creating these overrides is to ensure that critical data is transmitted to the correct location in real time, when the critical data is still useful.All the while, the device 105 maintains battery life by using a scheduler to only send less time-sensitive data.
[0034] In summary, data is sent in two situations: first, when data is scheduled to be sent and all data is shared ("regular delivery"), or second, when there is an override and specific data is sent to a specific location ("override delivery"). These overrides can be pre-programmed (by the user, the beneficiary of the information, the hospital, etc.) or these overrides can be determined by the device itself (based on learned triggers). In the case of dynamic triggers that learn the user's behavior and physiology, the dynamic triggers are determined by the application.
[0035] Smart communication stack decisions
[0036] According to one aspect of the present invention, the wearable device 105 is also configured to abstract the communication stack to provide support for multiple communication channels (such as Bluetooth, BLE, Wi-Fi, LTEm, LTE, etc.). This is extremely useful when the information is time-sensitive and the user cannot access their Wi-Fi or mobile phone. In such an instance, the device can use other available communication channels, including but not limited to LTEm. Based on pre-programmed rules downloaded to the device or learned by the device, the routing function performed by the routing engine receives outbound data marked by the device and inbound data, while the metadata indicates the urgency of the communication, the maximum "validity" (how long it can be buffered / delayed before it becomes irrelevant), the needs of downstream consumers of the data, etc. In one aspect, the metadata is configurable information that is transmitted from the cloud to the device and persists on the device until the cloud receives updated information from the device. The actual characteristics of the metadata (value, range, etc.) can be manually determined or can be calculated by some automated processing. The routing engine can also be configured to have a rule group that enumerates communication options, and the "cost" of activities such as battery impact, throughput, and other considerations for each communication option. The device can dynamically decide which communication stack to use at a given moment to increase speed, reduce cost, maintain battery life, etc. (for example, to transmit fall event data immediately without considering battery impact, and delay the use of higher cost options for low-importance, long-term use data (for example, data related to sleep periods) until lower cost options become available).
[0037] Current wearable devices use a default approach. The system 10 and wearable device 105 of the present invention have a smarter way to determine whether and how to send different types of data based on several considerations. For example, the device 105 can be configured to wait for Wi-Fi to become available rather than Bluetooth to send large data sets via Wi-Fi. The wearable device 105 can also be informed to make these decisions (e.g., based on battery power on the device 105). In the retail consumer goods arena, it is envisioned that each wearable device has a paired device. However, in different environments (e.g., hospitals, frail care settings, etc.), the envisioning is different.
[0038] Given that the wearable device 105 is configured to utilize multiple communication stacks, the wearable device 105 has several different communication channels on which the device 105 can convey / send information. In other words, the wearable device 105 is not limited to moving data in only a single direction. In addition, the wearable device 105 has the ability to update the device with new rules to handle communications, such as: under what circumstances to move data, how the algorithm samples, what data to sample and transmit, etc. The present invention may include an interface for a routing system to classify payloads based on priority, "validity period", etc. The communication channel can decide the best transmission to be used based on the "cost" of each transmission (battery, end-user data talk time, latency, throughput, etc.).
[0039] The method used by system 10 answers the following questions: (1) How is the decision to send a set of data made? and (2) Once the decision is made, over what channel should it be sent?
[0040] Current wearable devices use a traditional local storage approach with algorithms that collect and store information until the device 105 can communicate and send out the data. In order to dynamically control the behavior of the wearable device itself, the wearable device 105 includes a local information buffering system on the device that can store, retrieve, forward, and synchronize information as needed. This buffering system can move information between temporary storage and permanent storage in a way that improves the accuracy and consistency of the device 105, especially when the device loses power. For example, in order to calculate an accurate basic resting heart rate, about 1 to 2 days of heart rate data for a particular person is required. An accurate basic resting heart rate enables very accurate calorie expenditure calculations based on individuals. However, if the device loses power while still collecting information during the initial 1 to 2 day learning phase, an inaccurate resting heart rate and then inaccurate calorie calculations result. The buffering system allows this information (in temporary storage) to be backed up to permanent storage and restored if there is a power loss, thereby achieving improvements. The system and method of the present invention have expanded this to include functionality with a stored profile. The profile is not just a biometric of the user, but also includes algorithm information and state and device behavior configuration information.
[0041] For example, a pulse waveform algorithm using a set of rules according to which raw PPG data is collected and pulse waveform analysis is performed can be modified. If the system and / or device detects that something is abnormal or detects that something exceeds a threshold and needs to be investigated, the system can send update information to the algorithm profile to expand the scope of the pulse waveform algorithm for a specific individual / change the parameters of the pulse waveform algorithm for a specific individual / etc., thereby providing the ability to change the rules about how the algorithm operates. The standard pulse waveform algorithm requires a lower resolution raw PPG signal, and when a potential event of interest is identified, the resolution of this data can be temporarily increased, thereby providing the algorithm with an input with higher fidelity, which in turn produces an output with higher specificity and sensitivity. This is useful because using this method, firmware updates do not require improvements or changes to the algorithms stored on the device. In addition, such functionality supports multiple use cases. The traditional approach to multiple use cases has been to build firmware, each with a unique and fixed set of algorithms with fixed predetermined behaviors. Now, changing the algorithm profile for a group of devices or for an individual device changes the sampling behavior for a specific application.
[0042] The profile information can be divided into three categories: (1) human information, (2) algorithm information (e.g., configurable information that dictates behavior and modifies some algorithm parameters), and (3) device information (configurable device settings (e.g., whether and how the device should vibrate for certain events such as low battery or reduced sensor signal quality). Having this type of information stored locally on the device 105 and being able to modify it is novel in the art. This type of information should not be stored in random access memory (RAM) because it would be lost when the battery is depleted or the device is reused. Therefore, this buffering system is used to persist the information in non-volatile memory. As the device learns about the user, it can update the stored profile to improve accuracy. Although the role of buffering is known in the art, wearable devices use buffering in a unique way—saving human biometrics as well as configurable device settings.
[0043] The buffering system also saves the algorithm state, which is the stage / state that the algorithm is in at a given moment. In some aspects, these algorithms take a given amount of time to generate an accurate result; some algorithms may require hours or days of biometric data to converge on a final answer. For example, as discussed above, it may take hours and days to converge on an accurate basal resting heart rate for a particular person by observing their 1 Hz heart rate dynamics. Additionally, specific conditions may be necessary (e.g., sedentary conditions when the heart rate is stable) that may not be available for a period of time. Therefore, it is helpful to know where the algorithm is in its process of delivering a result.
[0044] An example application of this is that if you want to know the user's resting heart rate, the device can build a histogram of candidate information for the user's day. This histogram can record the user's heart rate and duration in a stable state at different times of the day. Over time, the histogram is filled. It is desirable to avoid running out of battery on the device and running out at noon, losing all this information. The buffer system allows the algorithm to aggregate information over time to request the buffer system to continue to retain / save its state (i.e., a certain moment in the process and the data collected / generated before that moment), which makes the buffer system useful. Therefore, when the battery is exhausted and the algorithm restarts, the algorithm can retrieve the stored state from the buffer system and restore the histogram to the last state it was in before the restart, and continue recording from where it was interrupted (in the instance of resting heart rate). For the algorithm, what is most useful is that it is necessary to understand / aggregate information over a period of time to produce a result, so it is important that the information that has been collected before the final result is produced is not lost.
[0045] Associating users and data sources
[0046] In the prior art retail consumer wearable space, there is an assumption that the same company maintains the roles of registrar (access to personally identifiable information), source (wearable device), transformer (party that transforms the collected information into new information and insights), and consumer (applications, portals, etc. that use the collected and transformed information to interact with the user) under the General Data Protection Regulation (GDPR). This is because current manufacturers are vertically integrated and all GDPR roles are performed by the same organization. However, once you start looking at integrating this technology into other areas that require specialized solutions (such as hospital data dashboards, remote monitoring, research, etc.), you go against the retail consumer wearable model assumption. In addition, the traditional mapping of user <-> device <-> user phone <-> device company cloud <-> device company data transformation can no longer be assumed.
[0047] The present invention takes a different approach. Registrars, sources, converters, and consumers are decoupled to avoid operating under the assumption that they are all the same organization. The present invention provides functionality on the device to accept this assumption. Thus, data can be separated from personally identifiable information (PII).
[0048] This is particularly useful when information must remain confidential to comply with privacy and licensing regulations (such as HIPAA guidelines). Figure 3) is an embodiment of the de-identification process, acting as an intermediary between the wearable device 105 and the cloud / core / application backend 102 / 103, separating the data and creating unique identifiers for user access information and for application access information. If the system is damaged or compromised, the data is anonymous and cannot be tied back to a specific user. The DBN 107 registers each new device 105 as a new "data source" and assigns the source a unique identification code. When different parties consume this information, it is passed using a new unique identifier. Translators (e.g., The DBN 107 is a DBN that is used to store data and data between two parties. ...
[0049] This approach also provides the ability to easily introduce new users and associate / disassociate that user with a specific device / data source, e.g. Figure 5 As shown. There is a paired device 406 associated with the wearable device 405, there are two potential locations where information is collected and the data source can be changed dynamically. This also applies to associating a user with a new device or using more than one wearable device at a given time. In the current prior art, a single user is "bound" to a single device (for example, John's Airpods are associated with John). This new method allows the device 405 to request a connection token 501 from the central site system 502 (see also Figure 3 ), thereby associating the device 405 itself with the new user. Then the central site 104 acts as a proxy for the DBN 107 to obtain a short-term connection code 503, and the device 406 retrieves the connection code 404 via the central site of the proxy DBN and registers the user. The following are two main applications of this user association method:
[0050] (a) Device pairing
[0051] A user may use more than one wearable device 105 at a time, and there is a "preference" as to which device 105 performs the most work at a given moment (this may be based on many factors such as battery life, type of work to be performed, etc.) The preference is based on predetermined rules.
[0052] (b) Equipment Exchange
[0053] When using this new method of transferring user data, the data is continuously flowing. However, note that paired devices can become swapped devices, such as there being one wearable device worn at night while sleeping, and there being another wearable device worn during the day.
[0054] Expanding on the functionality / utility of using multiple devices below. A master copy of the profile information held by the on-device buffering system is held in the cloud and is delivered to any new device associated with that user. In this way, it is easy to switch devices for a given user without losing profile information related to that particular user. This is unique compared to other wearable devices currently on the market, as other devices do not have the flexibility to quickly and seamlessly associate a new device with a new user without losing profile information. Note that a DBN is used here to ensure that devices / sources do not access a user's information during a change in data source; profile information is stored in association with the user. This allows devices to be associated to new users without the need for personally identifiable information.
[0055] Distributed Sensing
[0056] This wearable device 105 is not limited to a specific area of the body. Some examples of locations where the wearable device can be worn include: on the wrist, on the upper arm, on various locations on the torso, on the face (near the temple for goggles, near the in-ear area for ear products), etc. The device 105 can be combined with a variety of straps, vests, and other garments with pre-sewn insertion locations, clip-on / magnetized housings to attach to compression vests, underwear, shorts / trousers, etc. The device 105 is configured to mechanically / electrically identify the location of the device on the body based on where / how the device fits with the strap / strap / clip-on housing / magnet, etc.
[0057] If the device 105 can be moved between the housing / attachment mechanism, the device 105 is designed to detect that it has been incorporated into different positions in different ways. This is critical when measuring physiological data measurements, because pulse, heart rate, movement, etc. can be measured differently in different positions. Position detection can be accomplished using signal analysis to automatically detect the position of the device on the user. For example, a pulse signal can be collected from somewhere other than the wrist, and the signal can be analyzed (there is existing software for this) to determine where the pulse actually comes from. Then, the device 105 can change its formula based on the position of the device. This is useful because different positions and different signal strengths will not produce the same results when using the same formula and algorithm. Therefore, it is key to know the position of the device and the impact of the position on a specific sensor signal.
[0058] In another aspect, the system may also incorporate a network of sensors associated with the user that collect information on a single body at a given moment. In some variations of this, the network may employ disposable "remote sensors / accessories / peripherals" that appear as remote peripherals / extensions of a wearable host device that has maximum rechargeable batteries, memory, processing power / power, and communication options. The peripheral device may have a fixed button battery, a specific sensor with minimal electronics that maintains the sensor and buffers the data, and then only utilizes BLE / Bluetooth communications to download the data to the host device.
[0059] Such a setup can be utilized in the context of a cardiology solution. There are algorithms that can use normal PPG signals for analysis and highlight arrhythmias that are highly likely to occur. However, in addition to PPG data, a doctor may want to be able to get a week's worth of ECG recordings. The wearable host device can be configured to run with a disposable wearable ECG patch that can communicate with the wearable device via Bluetooth. Instead of utilizing an external product to collect data with the ECG device, the patch can utilize the wearable host device for further edge processing since it is a communication agent.
[0060] Another application of the master / peripheral system is a consultation solution. For example, in a remote environment (e.g., after an outpatient surgery), a doctor needs a stethoscope to get data about their patients. A digital stethoscope patch can be sent to a user, who can then place the digital stethoscope patch on their body. The patch (peripheral device) communicates digital stethoscope data to a wearable device (master device). The doctor can view / communicate with the wearable device to view the data and control the recorded measurements. In the current art, an external digital device such as a stethoscope will utilize its own independent communication method (such as being directly linked to a single machine), and the external digital device is not configured to communicate with a wearable device. This new method creates a new communication agent and allows the use of a wearable device system and connection to other communication devices.
[0061] This ability to integrate otherwise disposable or permanent devices into a system that can communicate with a wearable host device has many key benefits. Collecting ECG data, for example, can be extremely invasive and expensive; cardiologists typically hand patients high-friction ECG Holter devices to wear for days / weeks. The present invention provides a cheaper, easy-to-use way to access the same data remotely. In addition, the ability to add auxiliary devices significantly expands the functionality of the wearable host device. Wearable devices are "future-proof" because when new methods of data collection / sensing technology exist, new devices that utilize the new methods can simply be incorporated. For example, a wearable host device can collect data from digital scales, blood pressure cuffs, thermometers, etc., which can communicate wirelessly with the wearable device and use the wearable device as its agent.
[0062] Algorithm-based physiological state
[0063] In the current state of the art, the standard approach to collecting physiological measurements is to solve an algorithm for the ultimate goal and / or worst case scenario. The system measures heart rate for downstream applications. This is not possible with other currently available devices on the market, as continuous monitoring would drain the battery. Prior art devices (see Figure 1 ) measures the heart rate at arbitrary intervals (e.g., every fifteen minutes). This information is useless in a physiological measurement. When this "interval method" is used, the information data is ignored.
[0064] An improved embodiment of the present invention is when monitoring heart rate. In a twenty-four hour period, people spend most of the day sleeping or not moving at all. The user's activities can be divided into three physiological states: (1) no exercise (~40-60% of the average user's day), (2) a small amount of exercise (~40% of the average user's day), (3) a large amount of / training level exercise (average user's day <= 2%). The wearable main device 105 is able to distinguish between different states and apply different algorithms to each physiological state, creating different algorithms to optimize the signal quality, coverage, and power consumption for each situation. A large amount of exercise is the most intense physiological state and consumes the most battery, while no exercise is the most relaxing physiological state. This new method provides complete coverage of physiological measurement data and utilizes power consumption similar to or less than the duty cycle method and higher efficiency than the duty cycle method. The old method enables and disables algorithms, while the present invention continuously utilizes algorithms and switches them.
[0065] In one aspect, the wearable device utilizes a supervisory algorithm that runs separately from the three separate heart rate algorithms. The supervisory algorithm maintains awareness of the user's physiological state (e.g., sitting, light activity, exercising, sleeping, etc.) and switches between the three heart rate algorithms as the user switches states. For example, if the user switches from light activity to sedentary, the algorithm optimized for the sedentary state is activated by the supervisory algorithm. In one aspect, the supervisory algorithm can take a variety of forms and steps depending on the characteristics of the algorithm it is supervising, and can be configured to monitor the user's state and battery status depending on the needs of the user.
[0066] Algorithm layering / extraction decomposition
[0067] When dealing with more complex algorithms, the question is when complex processes can be divided into sequential stages. When algorithms are divided into multiple steps, the right approach allows removing the dependencies between the algorithm model being run and where the algorithm model can run.
[0068] For example, sleep monitoring can be broken down into four parts: (1) real-time sleep / wake monitoring, (2) raw feature aggregation, (3) model feature extraction, and (4) sleep architecture neural network, heuristic model. In the standard / traditional approach, data is collected on the device, sent to the cloud, and then split in the cloud. However, if split into four parts, the more critical parts (e.g., algorithms that have the task of accurately identifying potential sleep events and capturing the data required for sleep analysis (transmitted from the device for further processing)) can be determined, require less battery power, etc., and only these parts are run on the device. In one aspect, the determination of the split can be predetermined (i.e., determined manually during the design phase of the algorithm code, based on construction and deployment), or can be learned through machine learning. In addition, these discrete transformations can be distributed on different processing layers of a wearable device, multiple wearable devices, other edge devices, and / or the cloud. Such processing provides immediate information by executing functions on the wearable device and obtaining temporary readings and then performing deeper and more accurate analysis in the cloud by utilizing more computing resources. A suitable embodiment is when an individual wakes up late at night. Rough calculations (e.g., sleep quality, SpO2 levels, etc.) can be performed on the device while more accurate calculators in the cloud are used for more in-depth studies. The processing that can be done on a wearable device is limited, so being able to choose what device does what analysis where is useful.
[0069] Device pairing mechanism
[0070] The wearable master device can also communicate with a paired device. The wearable master device can be configured to collect information from a second device, such as a smaller device that the user prefers to wear when resting. It is also possible to wear two devices at the same time; the master device is able to determine which device should collect data at a given moment. This relationship allows the wearable master device to maximize efficiency by conserving battery power when there is less "work" to be done. There is a defined relationship between the wearable master device and another device. The user can choose for themselves which device to wear at a given moment, or wear both devices. One device may be more accurate than the other, and one device may be more comfortable, etc. Both devices can perform physiological detection. Some sensing functions on the two devices may be duplicated (different in quality, coverage, and battery consumption impact). Devices may also have their own unique, non-overlapping sensing functions.
[0071] Some examples of how this system can be used include:
[0072] (a) Use alternative devices that have overlapping functionality (sensing, computing, storage, etc.) that is enabled only on the one device with the highest ROI. In this example, the master device determines which device will be used.
[0073] (b) Using devices in a "relay" fashion. Meaning, using a small / comfortable device during sleep and a larger device during the day. Multiple devices together provide 24 / 7 coverage and data collection. The user decides which device will be used.
[0074] (c) The attached sensors on multiple wearable devices sample the signals at the most appropriate locations and the data streams are merged and sent to the cloud (multi-location collaborative sensing).
[0075] Two or more devices can be configured to sense each other (proximity sensing) using Near Field Communication (NFC). The present invention utilizes NFC functionality that can be used in conjunction with / instead of Bluetooth. If the Bluetooth signal is strong, the Bluetooth signal will be used to listen to other wearable devices, and the strength of the Bluetooth signal can be measured. By being able to measure the strength of the signal, you can begin to create a proximity log of a specific device to know where the device was at a given moment (note that Bluetooth requires less battery power than GPS). When you have all of this information in the middle, you now have contact tracing information as well as proximity location information. One embodiment that utilizes this feature is in a hospital environment (see Figure 2 ), which can track which nurse saw which patient at a given moment.
Claims
1. A system for continuously monitoring a user's physiological state, include: a. a cloud configured to monitor and analyze physiological data related to the user; b. A non-invasive instrument, which is used as a data acquisition device by the user, comprising: i. Power supply; ii. at least one sensor; iii. a wireless communication device configured to communicate with the cloud; iv. Memory; and v. at least one processor, wherein the processor is configured to: A. determining at least one biometric from the data collected by the sensor; B. communicating the at least one biometric to the cloud; C. determining when to send the at least one biometric to the cloud based on the state of the power source and a characteristic of the at least one biometric; Wherein the cloud is configured to process the at least one biometric from the data acquisition device to determine a current health status of the user.
2. The system of claim 1, wherein determining when to send the at least one biometric to the cloud follows a set of rules that can be pre-programmed on the non-invasive instrument or learned by the non-invasive instrument.
3. The system of claim 1, wherein the non-invasive instrument is capable of near field communication to transfer data from the first device to the second device.
4. The system of claim 1, wherein the non-invasive instrument is capable of communicating with a second nearby device. The system of claim 1 , wherein the system is configured to identify a source of sensor data.
6. The system of claim 5, wherein the non-invasive device is capable of determining a proximity relationship between more than one sensor data relative to each other, and determining a source of sensor data about the user.
7. The system of claim 1, wherein the non-intrusive device stores user profile information, include: a. User information; b. Algorithm information; and c. Device information.
8. The system of claim 7, wherein the system retrieves the user profile information and sends the data via a communication channel, the system determining the best communication channel for a particular communication based on one or more metrics, the metrics include: a. The status of the device; b. the battery power available on the device; c. The battery power required to complete the communication; d. the processing power of the device; and e. The processing power required to complete the communication.
9. The system of claim 7, wherein the algorithm information on the device allows the device to filter data to be sent and to be collected, further reducing battery consumption and ensuring sustainability.
10. The system of claim 1, wherein the system is further configured to detect abnormal values of the user's health and send the data directly to the cloud.
