Environmental sensing for care systems

Abstract

A system, apparatus, and method for monitoring at least one person in at least one environment. The environment includes at least two sensors capable of detecting the presence of a person in the environment. The monitored person has a care state that is monitored, and such monitoring includes the at least two sensors providing a data set to at least one signal monitoring system. Such data set can be communicated to at least one digital twin representing the monitored person and its environment to determine the person's behavioral patterns. Such patterns may be represented in the at least one digital twin to detect behaviors indicative of a change in the person's care state being monitored.

Description

[Technical field]

[0001]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Application No. 63 / 328,101, entitled “Environment Sensing For Care Systems,” filed April 6, 2022, the disclosure of which is incorporated by reference in its entirety for all purposes.

[0002] Aspects of the present disclosure generally relate to systems for monitoring people in care. [Background technology]

[0003]

[0003] In a conventional infrastructure technology environment, Personal Emergency Response Systems (PERS), also known as Medical Emergency Response Systems, allow a person to call for help in an emergency by pressing a button.

[0004]

[0004] One exemplary system is a two-way voice communication pendant that allows a person to call for assistance anywhere around the home. Personal emergency response devices offer the potential for aging in place and independent living for people in care. Personal emergency response devices allow a person to stay connected with family and emergency services via their existing landline phone. Summary of the Invention

[0005]

[0005] Embodiments include systems and methods for monitoring a care recipient.

[0006]

[0006] A system monitors a care recipient by at least one caregiver. The system includes a care analytics management processor (CAMP) and a plurality of environmental sensors. Each of the environmental sensors includes at least one elastic repository configured to store a dynamically configured amount of sensed data from the care recipient's environment. The environmental sensors are each connected to at least one computer-readable medium configured to store the sensed data generated by the environmental sensors. The environmental sensors are configured to sense the care recipient's environment, determine a quiescent state of the care recipient's environment, and detect an edge state that deviates from the quiescent state and results in edge state data. Upon detecting an edge state, the environmental sensor is configured to evaluate the sensed data and the edge state held in the at least one elastic repository to determine whether the environmental sensor changes an active state. Upon changing to an active state, the environmental sensor transmits a configuration specification to at least one other sensor of the plurality of environmental sensors proximate to the environmental sensor. The active environmental sensor transmits the sensed data and the edge state data to the care analytics management processor. The care analytics management processor further includes a transceiver and a microprocessor. The transceiver is configured to receive the sensing data and the edge condition data. The microprocessor is configured to determine if a false positive condition occurs. When a false positive condition occurs, the transceiver is configured to transmit configuration specifications for resetting the plurality of environmental sensors to a quiescent state. When a positive condition occurs, the care analysis management processor is configured to transmit an alert to a caregiver.

[0007] In some embodiments, the transmitted configuration specifications are based on edge conditions detected by an environmental sensor.

[0008] In some embodiments, the transmitted configuration specifications cause at least one other sensor to dynamically invoke a threshold condition for detecting an edge condition.

[0009] In some embodiments, the transmitted configuration specification causes at least one other sensor to evaluate the elastic repository to determine whether the other sensor has changed to an active state.

[0010]

[0010] In some embodiments, the transmitted configuration specification causes at least one other sensor to evaluate the elastic repository and the detected edge condition to determine whether the other sensor has changed to an active state.

[0011] In some embodiments, the proximity to the environmental sensor is a physical distance.

[0012] In some embodiments, the proximity to the environmental sensor is a logical distance.

[0013] In some embodiments, proximity to the environmental sensor is based on line of sight.

[0014]

[0014] In some embodiments, the stationary state is based on the person's physical activity level.

[0015] In some embodiments, the alert is a call, text message, or electronic message to a caregiver or emergency services.

[0016] In some embodiments, the elastic repository records at least 30 seconds of prior data. In some embodiments, the elastic repository records at least 5 minutes of prior data. In some embodiments, the elastic repository records at least 1 hour of prior data.

[0017]

[0017] In some embodiments, the care analysis management processor is a wearable sensor configured to be worn by the person.

[0018] In some embodiments, the care analytics management processor determines that a false negative has occurred by analyzing the sensed data and edge condition data from two or more environmental sensors.

[0019] In some embodiments, at least one of the plurality of environmental sensors is an active emitting sensor device configured to create a map of the environment. In some embodiments, the map of the environment is a three-dimensional model. In some embodiments, the active emitting sensor device is a radar device, a light detection and ranging (LIDAR) sensor, a radio frequency (RF) sensor, or a frequency modulated continuous wave (FMCW) radar sensor.

[0020]

[0020] In some embodiments, the care analysis management processor is further configured to maintain an up-to-date and accurate positioning of the object designated by the person.

[0021] In some embodiments, at least one of the plurality of environmental sensors is a microphone, a camera, a strain gauge for shock detection, a thermal sensor, a motion detector, or a tactile sensor.

[0022]

[0022] These and other aspects will be described in detail with reference to the following drawings. [Brief description of the drawings]

[0023] [Figure 1] FIG. 1 shows a sensor-enabled environment (102) in which a monitored person (PUM-101) resides. [Diagram 2] FIG. 1 illustrates an exemplary set of modules that combine to provide a system for monitoring a care recipient. [Diagram 3] FIG. 1 illustrates an environment having a PUM (101) having one or more attached and / or wearable devices and / or sensors (104) within an environment (102) that includes one or more additional sensors (103). [Figure 4]FIG. 1 illustrates a PUM with a set of attached wearable devices. [Diagram 5] FIG. 1 illustrates a PUM (101) habitat that includes one or more sets of sensors. [Figure 6] FIG. 1 illustrates one or more sensors configured as edge devices for monitoring PUM in an environment. [Figure 7] FIG. 1 illustrates state management by a care analysis management processor (105) for an environment (102) having a PUM (101). [Figure 8] FIG. 8 shows an HCP (801) in operation. [Figure 9] FIG. 1 illustrates several PUMs (901, 911, 921), each existing in a different environment (902, 912, 922). [Figure 10] A diagram showing the use of a care village digital twin (701). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0024]

[0033] Aspects of the present disclosure include systems and methods, including devices and methods, for monitoring a care recipient.

[0025]

[0034] There are many sensors that can be applied to an environment to determine activity within the environment. These include sensors that actively transmit signals into the environment, sensors that capture photons or other electromagnetic frequencies from the environment, sensors that capture acoustic and other air pressure signals from the environment, sensors that capture movement in any direction, all of which are portable and can be carried into the environment. There are devices that incorporate various combinations of these sensors, such as accelerometers, gyroscopes, altimeters, etc.

[0026]

[0035] FIG. 1 illustrates a sensor-enabled environment (102) inhabited by a monitored person (PUM-101). Data sets generated by such implantable sensors (103) and / or portable or wearable sensors (104) are communicated to one or more care analytics management processors (CAMPs) (105) that can determine behavioral patterns (107) represented by that data. These patterns include, at least in part, representations of the PUMs and their health care profiles (HCP-106), enabling detection of their activities within the monitored environment that supports detection and identification of one or more changes to those patterns that are indicative of a change in the care and / or health of the PUMs. This data, patterns, configurations, specifications and any other management information can be stored and / or communicated to one or more care village digital twins (701).

[0027]

[0036] In some embodiments, the HCP includes a framework that includes a specification of an initial configuration of one or more sensors to be used to monitor a person designated by the HCP. This can include one or more specific care and / or health related behaviors, activities and / or events, individually and / or in patterns. For example, a PUM with memory impairment can have a specific sensor configured to identify repetitive behaviors indicative of such impairment, including monitoring fluctuations in the degree of such impairment. In another example, if the PUM has an HCP that includes a specification regarding the likelihood of falling, one or more sensors can be configured to identify behaviors that indicate an increased likelihood of that event occurring. The HCP can include patterns that represent the monitored behavior of the PUM, and changes from such behavior can provide an indication (in some embodiments, at least in part determined by one or more machine learning techniques) regarding the likelihood or potential of a health or care event. In such situations, the sensor configuration can be modified to configure the monitoring to accurately provide data to one or more responding parties and / or systems. This dynamic configuration of one or more sensors, devices and / or systems can provide further data that identifies changes in the PUM's behavioral patterns, which may change the configuration of one or more of the sensors, devices and / or systems, including, for example, providing such data to one or more parties involved in providing care and well-being for the PUM.

[0028]

[0037] One aspect of the data set generated by a sensing device can be classified as either constant or intermittent. Constant sensing, such as with a thermometer or barometer, simply captures the information produced by the sensing function and displays, stores and / or transmits this information. Intermittent sensing occurs in response to an activity, event, timer, or other occurrence, such as a human request or interaction, a trigger, periodic timer or other event activation, where the sensor data is directed to another device or system and / or the sensor itself is activated.

[0029]

[0038] A further classification is that of the sensors themselves, which can be either active or passive, in that the sensor either produces one or more emissions or does not produce any emissions.

[0030]

[0039] Sensors can collect, measure, process, store, and transmit data in any combination depending on the sensor's capabilities and the configuration used. In some embodiments, sensors can measure, store, and / or transmit data based on the sensor's capabilities and configuration. Sensors may be collecting and / or measuring or not, as determined by the state of the sensor, and can be configured by the system. In some situations where a sensor is measuring, it can be configured not to store or transmit data unless or until a command, threshold, or other trigger, action, or event, including a time, is received by the sensor and / or the system controlling such sensor. For example, a sensor can not measure, store, and transmit data, or can be configured to transmit data on demand, such as by specifications held by the device or upon receiving a command from the system.

[0031]

[0040] In some embodiments, sensors, devices, and / or systems may use one or more elastic repositories, where the storage available to the sensors, devices, and / or systems may be dynamically adjusted from a minimum capacity to a capacity determined by one or more configurations deployed for the sensors, devices, and / or systems. For example, a sensor will likely have a fixed amount of storage built into the sensor. This capacity may provide, for example, the ability to continuously store up to an amount of data equal to a period of time at a particular resolution.

[0032]

[0041] The configuration of a sensor may include access to one or more additional repositories connected to the sensor via one or more communication methods. For example, if the sensor is connected to a dedicated hub, router, or other device configured to accept such data from the sensor, the dedicated device may provide an elastic repository that can dynamically provide additional storage capacity to the sensor. In some embodiments, the dedicated device is a care hub.

[0033]

[0042] The determination of available repository capacity can be represented by one or more sets of configuration specifications that may be determined in whole or in part by the HCP's motion patterns. For example, if the motion pattern state is stationary, the storage capacity provided may be configured to exceed the storage capacity available at the sensor itself. This capability may be configured by a care hub, router, or other dedicated device to be a certain size of data and / or a certain resolution length of time. In this manner, a combination of sensors, devices, and / or systems connected to one or more dedicated devices including a care hub, router, etc., can provide a data set that informs about a condition before an event or alert of a care or health event is generated by, for example, an edge device.

[0034]

[0043] For example, if a PUM trips over a pet or piece of furniture, certain sensors may detect a fall, but the situation is just a minor trip, and therefore pre-event data may confirm that this is the situation, thus avoiding a false positive and potentially unnecessary response.

[0035]

[0044] For any one or more sensors, there is a quiescent state from the perspective of the system monitoring the environment, where the sensor is not providing data to the system or there is no change in its data. A sensor can be operational and have at least one of the following states: collecting, measuring, processing, storing and / or transmitting data to a system that has configured the sensor and established command and control of the sensor operation.

[0036]

[0045] FIG. 3 shows an environment having a PUM (101) having one or more attached and / or wearable devices and / or sensors (104) within an environment (102) including one or more further sensors (103), all of which are configured and partially or wholly managed by one or more care analysis management processors (105), whereby the system establishes a quiescent state (109) of that PUM within the environment in the context of HCPs (106) including one or more behavioral patterns (107), one or more of which are operating (108).

[0037]

[0046] In an environment having one or more sensors, a quiescent state can be established by the system using an appropriate care analytics management processor or processors, which may be determined from the data and / or may be established over a period of time during which data from the sensors are continuously consistent with little or no variance.

[0038]

[0047] In an environment with multiple sensors, each of these states of the sensors may be configured, in part, as part of a system for establishing a quiescent state of a monitored environment including several sensors. This may include one or more sensors measuring and sensing the environment. For example, the transmission of data from the sensors or set of sensors to other devices and / or systems may not occur individually or collectively based on the state of the sensors and the environment. This may include the sensors being inactive and dormant, or the sensors being configured to transmit only data related to events, triggers, thresholds, and / or actions, for example, generated externally by the system or derived from the measurement and processing capabilities of the sensors. The combination of the states of the individual sensors may be integrated such that a care analytics management processor, which may incorporate one or more sets of command and control functions, may configure their operational state and / or manage which sensors can communicate with other sensors to change their operational state.

[0039]

[0048] Sensors may be integrated into devices. For example, a typical smartphone may have multiple sensors embedded within it. In other examples, a sensor may be a standalone device with a single function, such as a micro-electromechanical system (MEMS) microphone intended to capture acoustic signals. Such standalone and / or single-function sensors may be aggregated to form a sensor set with greater capabilities than a single sensor.

[0040]

[0049] In some embodiments, sensors can capture data from the environment and such data can be stored by the sensors. This can include sensors having sufficient on-board storage capabilities and / or access to a repository suitable for such data storage. This data can be stored on any type of basis, such as FIFO, where data representing a particular length of time, period of a day, amount of data or other metric is stored in the repository. Data sets from any sensor can be configured to be stored for a period of time, up to an amount of data, until an event or message is received, until further configuration specifications are deployed, and / or until other conditions and / or specifications are invoked.

[0041]

[0050] In this way, a sensor can hold a set of data that can be accessed when an event occurs to ascertain the conditions represented by that data prior to the event. As multiple sensors may be placed in an environment, data from such sensors, when combined, can provide valuable insight and context to an event. For example, this can be particularly useful in detecting false positives or providing data that can assist with emergency or other responses. For example, difficulty in breathing may be detected by a combination of sensors prior to a fall potentially indicating the cause of the fall.

[0042]

[0051] In the case of passive sensors, they continuously collect data and store it for a configured predetermined time, e.g., 15 minutes, after which the data is deleted from the sensor repository. If the sensor is active, i.e., uses emissions to create a data set, such as FMCW radar, these sensors may be configured to perform periodic emissions as part of an operational monitoring pattern. The data so generated may be stored in the repository in a similar manner as passive sensors.

[0043]

[0052] In some embodiments, there may be one or more devices described herein as edge devices that can initiate a change in the state of at least one other device, e.g., from a quiescent state to an active state, to another state. Such initiation may be initiated via a sensed event, a timed and / or configured arrangement, and / or via a pattern specification that includes one or more such events. These initiations may include a common system that integrates the sensors, such as a command and control system, and / or may use direct communication between the sensors, e.g., based on appropriate protocols and security schemes.

[0044]

[0053] In some embodiments, an edge device may be a device set including at least two devices that generate an event signal that triggers a processing step that changes the state of at least one other device. This processing may occur on the devices and / or on one or more servers in any configuration.

[0045]

[0054] In this manner, the specification of a pattern in combination with one or more environmental sensing devices may be combined to change the state of one or more other sensing devices, which state change may include changing the configuration of the sensing device to focus the capabilities of the device on a particular event type, behavioral pattern, and / or other focus.

[0046]

[0055] In some embodiments, combinations of sensor data can be created by the extension of a single sensor with the capabilities of another sensor. For example, an acoustic sensor, such as a MEMS microphone attached to a hard surface such as a window, can have a frequency response configured for detection of low frequencies, such as footsteps of a person under surveillance (PUM), and can therefore collect, measure and process the detections and create an event. For example, an initial sensor can detect an acoustic signal that matches a pattern designated as a "trip," where, for example, the PUM has missed a step, which can indicate that the PUM is unstable and is about to fall or is in the process of falling. Such an event can then trigger other devices, such as smart speakers, smartphones, smart TVs, etc., to turn on their microphones, screens, speakers or other capabilities and, if capabilities exist, provide stored information, such as information buffered or cached prior to the event captured by the first sensor that created the event, which can then be combined by the system to create a richer data set for analysis by the care analytics management processor.

[0047]

[0056] Additionally, some sensors may have relationships with other sensors, such relationships being pre-configured in a configuration such that upon an event detected or triggered, either directly or via communication methods, the sensors provide data sets to each other and / or to the monitoring system. For example, when a sensor is triggered or detects an event, the other sensors may provide a data set that enables the initial sensor or any sensor in such a configuration to take on and / or provide additional functionality suitable for enhancing, expanding and / or extending the data set that may be provided to one or more care analysis management processors.

[0048]

[0057] In an example where one of the sensors in the configuration is configured as an edge device, this allows other devices in the configuration to change their configuration in response to edge device communications. Such communications may change in accord with edge device sensing capabilities such that upon detection of a particular type of event, which may be specified as a pattern, stored locally on the sensor or host device or in a repository available to the sensor or host device, the edge sensor may send a different set of communications including instructions for the configuration of other sensors. In some embodiments, such communications may be in the form of a token. The ability of a sensor and / or device to host or connect a first sensor to another sensor in its vicinity to send configuration data to that sensor allows the set of sensors to adaptively respond to changes in the PUM and / or conditions of the environment they currently occupy.

[0049]

[0058] In some environments, there may be sensors designated as dedicated edge sensors configured to capture events that indicate the state of the environment has changed from a quiescent state to an active monitoring state. For example, this may be a motion detector, an acoustic detector, a camera, etc. Edge detectors so configured may have configurations that respect the privacy of the PUM. For example, if the PUM uses the bathroom, a camera may be an inappropriate sensor, and a motion detector or a MEMS microphone configured to detect a foot drop, or a strain gauge configured to detect floor pressure, may be more appropriate and provide an initial signal of that state change. In some embodiments, the sensors may incorporate one or more event data sets representative of such events and / or may be configured to be triggered by one or more events that exceed one or more thresholds or other configuration specifications.

[0050]

[0059] 6 illustrates one or more sensors configured as edge devices for monitoring a PUM in an environment. A PUM (101) in an environment (102) includes one or more sensors (103 / 104), one of which is designated as an edge sensor (110), which provide data to one or more care analytics management processors (CAMPs) (105). In this illustrative example, the care analytics management processor (105) includes sensor monitoring (112), sensor configuration (113), monitoring focus (111), and state management (114), all of which may be embodied as modules of one or more care analytics management processors. The state management module may instantiate and / or store and / or retrieve a representation of the resting state (109) of the monitored environment. The care analytics management processor may communicate with an HCP (106), which includes behavioral patterns (107) and movement patterns (108) of any configuration.

[0051]

[0060] This can result in the deployment of edge devices that support monitoring of the PUM to instantiate behavioral patterns (107) for the PUM without violating the privacy of the PUM.

[0052]

[0061] In many situations, it is possible to generate false positives, where a single device detects an event, e.g., the sound of something falling, which can trigger a response. It is highly beneficial to use edge sensors to interact with other sensors, devices, and / or systems to confirm the accuracy of the situation. This can be done by activation of different sensors, e.g., cameras, microphones, and / or other devices. The generated data can then be combined with the initial edge response and evaluated by the monitoring system and / or a human operator. In some embodiments, the initial evaluation can be done by comparing the individual and combined responses of the sensors to patterns indicative of the type of situation, e.g., stored as a pattern or expected occurrence of the person in the environment. This can include, for example, patterns indicative of a person falling, missteps, breathing abnormalities, breakages, or other occurrences that collectively and individually indicate that the PUM has or is likely to have an event that affects its care and / or health.

[0053]

[0062] In some embodiments, devices capable of two-way communication, such as mobile phones, smart speakers, smart TVs, etc., may be configured to initiate a dialogue with a person occupying the environment. In this manner, the person may communicate with a system that monitors the environment, which may include ML capabilities for evaluating the communication and / or may include one or more human monitors of the environment. For example, a device capable of generating an audio signal may ask the PUM a question such as "Do you understand?" and then listen to the PUM's response, either using the device or other sensors or devices in the environment. This response may be evaluated by a voice pattern recognition system to ascertain the degree of stress in the PUM's response, which may then trigger a further system response, such as a caregiver calling, messaging, and / or visiting the PUM.

[0054]

[0063] The creation of a pattern from a sensor integrated into the environment can include feature extraction as a subset thereof, allowing for the expansion of the sensor's features, where the sensor's capabilities can be expanded and / or augmented by integration with other sensors in the vicinity and / or by the expansion of the configuration specifications of that sensor as specified by one or more patterns. This can include the use of attribution values ​​and weightings of certain features that are being evaluated by the combined context rather than the sensor itself. This expanded and / or augmented data set can be integrated into the pattern and / or context. There can be one or more thresholds that can be dynamically invoked, managed by the sensor directly and / or in combination with other systems. These thresholds can then trigger the expansion and / or augmentation of the sensor, including the combination of other sensors in the vicinity. This allows the evaluation of features extracted from a single data set, for example an image recognition system, to be incorporated into a pattern that more accurately represents the situation occurring in the monitored environment. This can reduce the occurrence of false positives or negatives that are often caused by evaluating a single data set in isolation from the context in which it occurs.

[0055]

[0064] While there has been significant development in the field of autonomy, especially in the area of ​​mobility, the care village environmental sensing system, which is primarily self-adaptive, dynamic and responsive, is not intended to be fully autonomous initially. The care village system is intended to incorporate one or more autonomous subsystems, e.g., a robotic medicine dispenser, and these specialized functional systems will become part of the monitored environment. As autonomy becomes more prevalent and trustworthy, the integration of autonomous subsystems is likely to increase without adversely affecting the health of the PUM.

[0056]

[0065] FIG. 2 illustrates an exemplary set of modules that combine to at least partially provide a system for monitoring a care recipient as described herein. The HCP (106) communicates with the environment (102) including environmental characteristics (201), a profile of the PUM (211), and other participant profiles (202). This may include an environment framework (212). The system may include one or more sets of patterns, including one or more pattern frameworks (204), behavioral patterns (107), and movement patterns (108), that may be matched to the environment (102) and its elements using a matching system (203). The care analysis management processor (105) may communicate with all these system elements and may include one or more sensor data repositories (206), feature extraction systems (207), event sets (208), event sequences (209), and pattern elements (210) in any configuration.

[0057]

[0066] Environmental Specifications In some embodiments, an initial survey of the environment is conducted to ascertain the characteristics of the environment, including physical attributes such as dimensions, entrances, windows, floor coverings, furniture (fixed and movable), and includes the environment's amenities, e.g., electrical outlets, plumbing, climate control, etc. The survey may also include an inventory of the environment's furnishings and their locations within the environment, including areas of the environment that have specific purposes, such as bathrooms, kitchens, bedrooms, etc. This survey may be conducted in person and / or may be based on architectural or other existing plans. This survey may also include devices that can sense the environment.

[0058]

[0067] In some embodiments, a set of sensors, including those within the environment, can be used as part of the survey to create an environmental dataset that incorporates multiple perspectives from a range of sensors with different capabilities. For example, cameras or other photon capture devices can provide one perspective at visible light wavelengths, audio transmission and reception sensors and devices can provide an additional perspective dataset, and FMCW or other RADAR or LIDAR sensors and devices can provide an additional dataset.

[0059]

[0068] These data sets are then integrated to form a comprehensive survey and / or mapping of the environment, establishing boundaries and contents in a manner usable by the care village system. This can include multiple passes with different sets of sensors to establish the location, physical properties and other aspects of the environment and its contents with high accuracy.

[0060]

[0069] The sensors may be interrogated in real-time for additional validation and accuracy, or may be used to update the survey if a change in the environment is detected.

[0061]

[0070] These environmental survey characteristics are then stored by the system in one or more repositories as a specification of the characteristics of that environment. Initially, this information is stored as an environmental framework. Such a specification may be used by the system to calculate various attributes of the environment, such as potential acoustic fingerprints of the environment, temperature, humidity or other aspects of the environment at different times and seasons. This may include, for example, the type of floor covering, the amount and location of soft furniture, and the ratio of hard to soft surfaces. The physical characteristics of the person or persons occupying the environment also factor into establishing these characteristics.

[0062]

[0071] The framework can provide an initial representation of the environment and incorporate them as part of an environmental framework (ENFW) in environments with multi-room and / or limited outdoor space. This initial ENFW can then incorporate data sets over time from sensors embedded in and / or traversing the environment. One aspect of this is to establish a quiescent state of the environment, which can inform further evaluation of occurrences within the environment and, in some embodiments, can be thought of as a "noise floor" against which occurrences detected by sensors can be evaluated.

[0063]

[0072] In some embodiments, active devices that may include one or more sensors may be used in the environment to ascertain certain characteristics of the environment, for example, wideband acoustic transmitters may be used to determine the acoustic characteristics of the environment. Such devices may also be used to determine common acoustic signals for the environment, such as doors or windows opening and closing, chairs moving, etc. These factors may also be calculated using machine learning techniques to form a signature that is used in appropriate signal processing to increase the efficiency of that processing. For example, the detection of any signals commensurate with a change in the state of the environment may be enhanced by the elimination of potential false positives due to the activation of additional sensors and / or changes in the acoustic characteristics of the environment.

[0064]

[0073] Similar techniques can be used for other frequency and time domain characteristics, e.g., visual, infrared and ultraviolet wavelengths, RF wavelengths, RADAR, LIDAR, etc. This can include using lasers or other forms of coherent wavefront technology to establish baseline characteristics of the environment, e.g., walls or other boundaries to measure distances, etc. This can include identifying differences in surface height, such as differences in stair or floor heights. Such characteristics can then be classified as trip hazards or other potential aspects that may affect the well-being of the PUM.

[0065]

[0074] An inventory of equipment such as appliances, refrigerators, room cleaners or other autonomous, semi-autonomous and manually operated equipment may also be stored as part of the environmental specification. In some embodiments, the system may use general characteristics of such devices as placeholders in the specification and test these as hypotheses against the actual sensed information that is monitored to form a more accurate representation of the environment. When available, specifications of actual devices may be incorporated into the ENFW and may include, for example, power or water consumption, operating hours, and / or other directly or indirectly observable characteristics of the devices.

[0066]

[0075] Identification information There is a set of identities that are instantiated and deployed by the system for the provision and management of care to one or more persons. The identities are specific to the system and in some embodiments may use a distributed ledger and / or other immutable record system. The system may assign an identity to each sensor, device, environment, participant or other care village entity in any arrangement. These identities may be instantiated as immutable identifiers, i.e., representations of identities by identifiers. In the case of humans, this may incorporate, for example, biometric and other person-specific identifying characteristics such that the identification of a PUM represented by that identifier can only be met by that particular PUM if evaluated by one or more systems. Specific characteristics and / or other metadata, including context, location-specific, time, etc., may be used in conjunction with identifiers of care village entities such as sensors, devices, environments, etc.

[0067]

[0076] In some embodiments, a PUM with system-issued identity information may have relationships with one or more other system elements, such as environments, parties, devices, tokens, and other system entities and / or their respective digital twins. This relationship may be implicit or explicit, and thus form the identity context of that PUM. Such approaches may be used by the care village system to establish further identity characteristics, including those of a dynamic nature, for example, when a role, such as a caregiver, is being played, or when an interaction, invocation, and / or engagement is episodic by another party.

[0068]

[0077] Such a context may be embodied, for example, as a matrix or other form including a graph, vector, or other spatial and / or temporal representation, where each of the identities has at least one relationship with another identity. For example, this may include the PUMs and their respective environments, as well as sensors and devices embedded in the environments, and / or devices that the PUMs may carry with them. This may also include relationships with other people, such as family, friends, and / or caregivers or aides.

[0069]

[0078] In some embodiments, these relationships may include specifications that at least in part determine the use, distribution, privacy, processing and / or other system functions and capabilities by the care village system and / or identities therein, including functions and / or configurations that may be applied to these relationships. Such configurations may be dynamically applied in anticipation of and / or in response to one or more data sets provided by one or more sensors, devices, and / or systems.

[0070]

[0079] The system creates and deploys one or more identities that may be immutable to the environment. This identity may be in the form of a token that may include a cryptographic representation and may be stored in at least one distributed ledger and / or other storage system. Such an identity may include an organization of the elements that comprise that identity expressed as an identifier, such as a hierarchical or other arrangement that may include different regions or elements. For example, this may include separate identifiers for different rooms in a multi-room environment, etc. In some embodiments, devices and physical artifacts of their environment may be given identities that may be arranged in a hierarchical or other organization.

[0071]

[0080] The structure of these identity configurations can form a schema used by one or more services and / or systems when interacting with those identities. In some embodiments, such a schema can be dynamically created and deployed over a set of identities in response to and / or in anticipation of one or more state changes involving at least one of those identities.

[0072]

[0081] Stakeholder specifications. There may be any number of stakeholders that may participate in the care village. This may include one or more stakeholders including the person under supervision (PUM). The stakeholders may include types of roles such as individuals, organizations, groups of individuals and / or organizations, roles, classes, and / or functions. There may be multiple organizations of stakeholders, and each stakeholder has relationships with other stakeholders, for example as an ontology or taxonomy. For example, each of these PUMs is likely to have a set of other stakeholders with whom it has relationships, which may include, for example, professional relationships such as doctors, dentists, physiotherapists, caregivers, and other relationships such as family, friends, neighbors, etc.

[0073]

[0082] Each environment has at least one or more people, participants, who reside primarily in that environment, either permanently or temporarily. This participant is the monitored person PUM. There may be multiple PUMs, e.g., couples, residing in a single environment. Each of these participants has a set of characteristics including physical attributes such as height, weight, gait, and other characteristics that form the basis of the system stored biometric specification for that person. This may include other biometric and / or care-related characteristics, including previously recorded health conditions, such as blood type, eye color, hair color, etc.

[0074]

[0083] For example, if a PUM is at or under the control of a care village system for health and / or health monitoring, their confidential health records may also be made available to the system, in whole or in part. Such availability may be subject to one or more enforced privacy or security regimes. In some embodiments, this may include the use of a secure token. In many situations, a particular health or health condition is the initial reason that the PUM is being monitored by the system. This condition, described as the primary condition, may at least in part determine the configuration of the monitoring device for the PUM. However, as is often the case, the primary condition may have other health and care-related aspects that may affect the PUM, and thus the system may be configured to monitor any behaviors and patterns that have a care or health impact.

[0075]

[0084] There is a system initialization set of processes where the identification of the PUM and its environment, actors including their primary conditions and / or other health or care related conditions are created as system elements. This forms part of the Healthcare Profile (HCP) of the PUM.

[0076]

[0085] Each of these participants can have relationships with people and / or environments that have certain characteristics, some of which form patterns of behavior and interaction that can be used by the system to provide care monitoring.

[0077]

[0086] These relationships may be expressed in the form of bilateral or multilateral specifications, where the exchange of information between both parties is governed by these specifications. In many embodiments, this includes device-to-device, sensor-to-sensor, sensor data exchanges, where the is a mutual agreement, which may be binding, including, for example, smart contracts between devices. In some situations, these may act as representatives of the parties, with one or more sets of specifications governing the extent and / or content of data exchanged or expected to be exchanged. The exchange may be specification and / or context dependent and / or influenced by the current state of the environment and one or more people therein.

[0078]

[0087] Because PUMs are effectively the focal point of the care system, the extent of the data on their current status is likely to be more detailed, especially with regard to their health and wellness. This health and wellness data is likely the reason for them being placed under surveillance in the first place.

[0079]

[0088] Stakeholder Profile Each party in the system is issued a system-specific identity. This identity can provide multiple relationships with other parties, sensors, devices, environments, care systems, etc., in any arrangement. Common techniques such as the use of NFTs or other cryptographic tokens can be used for both the security, authenticity, and integrity of the identity and the privacy of the individual represented by such identity. These identities can include multi-factor authentication, such as typical two-factor, biometric multi-factor, location and other contextual data, thereby ensuring that each party is accurately and reliably identified to protect both the party and the care village from false or inaccurate representations of identity, such as typical phishing or other false identity representations. This includes minimizing the burden on the PUM or other party to establish, verify, and authenticate its identity.

[0080]

[0089] Such an identity system may use one or more biometric features as part of an identity representation, which may include fingerprints, retina scans, facial recognition, gait analysis, palm recognition, and / or other biometric data in any combination. This may include one or more sensors and / or devices similarly identified via specific features and identifiers of those sensors and / or devices, including confirmation of the relationship of the PUM to those sensors and / or devices, which may include biometric data, as a representative of the person for interactions with one or more care village systems. This may include multiple devices, each authenticated to one or more care village systems and / or other devices, and having co-location, proximity (physical or logical) and / or other characteristics that may be verified by one or more other sensors, devices, parties and / or care village systems, forming a set of data represented as a profile. For example, a PUM may have a PERS worn, and the worn state may be confirmed by the PERS and / or via other sensors and / or devices. The PERS may be a separate device and / or a smartphone configured to function as a PERS via the smartphone's on-board sensors. These devices, PERS and / or smartphones, separately or together, can communicate, e.g., using Bluetooth, to provide sets of data, e.g., in the form of one or more tokens, to a third device, e.g., a router capable of receiving Bluetooth communications, which then communicates with one or more care village systems.

[0081]

[0090] Such identities may be represented in one or more care village digital twins (CVDTs) and may be used by the care village system as part of the assessment and analysis of the PUMs and their environments. In this manner, CVDT assessments, modeling and / or predictions may include, for example, changes in the cadence of participants' behavioral patterns to more accurately represent the behavior of the PUMs.

[0082]

[0091] Actors can be actively or passively involved in the patterns and / or response configurations that are or may be unfolding in the environment, which may involve multiple, sometimes complex relationships between individual actors and their influence on the PUM.

[0083]

[0092] One aspect of the system is the ability to provide verification that certain care and health related activities have occurred, such as scheduled or prescribed visits by care and health participants, and the system acting on behalf of those participants. This can include devices dispensing prescription medications, etc. For example, this can include regular PUM self-care such as showering, regular sleep, exercise, and taking prescription or other medications in a specified manner. Many of these actions can be identified directly and / or by inference by at least one sensor to verify that the action occurred, verifying the timing, context, and other data as necessary. One verification technique is the use of proximity, i.e., the caregiver and PUM are co-located in the environment. This can include device-to-device communication using, for example, Bluetooth, with distance calculated through monitoring of signal strength. The use of near-field technologies such as NFC and / or RFID can also be deployed, as well as line-of-sight and / or other directional proximity technologies. Near-field wireless technologies can include the use of smart cards or other near-field wireless enabled devices, where the device and / or card must be touched to a reader for verification. For example, the environment may have fixed terminals for such transactions and / or the PERS may be configured as mobile terminals for this purpose. In some embodiments, detecting and / or verifying the presence of one or more parties may be sufficient, although in some circumstances, more stringent verification of the co-location of one or more parties may be required. For example, the use of cameras, radar, heat or thermal detection, etc. may be used to verify, verify or confirm the proximity of one or more parties to another party, e.g., a PUM. In some embodiments, one or more devices may use a dynamic QR code or similar bar code and / or other visual indication to ensure the proximity of one or more devices.These devices may be bound to their respective owners, for example to verify the owner's presence via fingerprint, camera or other biometrics, and to generate and read such indicia within a specified time frame.

[0084]

[0093] This approach may include the use of one or more biometric technologies capable of verifying the presence of one or more parties, which may include the use of short-range communications such as Bluetooth, short-range radio and / or high-frequency RF communications such as 60 Ghz, for example, using smart devices and / or PERS configured for such short-range communications.

[0085]

[0094] Such verification may include one or more sensors providing a data set indicating that such an event is taking place, such as a shower being flushed, a toilet being flushed, etc. Such an indication may be detected by an audio sensor for an initial indication and then confirmed, for example, by an increase in measured water flow. Each of these activities has a distinct profile in terms of both audio and water usage, thereby allowing detection of such events with a high degree of confidence. This data may be used, in whole or in part, as part of one or more patterns indicative of the behavior of one or more participants in one or more environments. Monitoring of services such as water, gas, electricity, internet, etc. may be integrated into the participant behavior patterns.

[0086]

[0095] In some situations, specialized devices and / or applications may be used to dispense prescribed or other medications. These devices and / or applications may be integrated into the environment, and the datasets generated by such devices may be complemented by additional datasets from other embedded sensors. For example, detection of water flow or the opening of a refrigerator door, for example using a MEMS microphone or other acoustic and / or other sensors, may add accuracy and reliability to a dataset indicating that a PUM has taken their medication at a particular time.

[0087]

[0096] The use of multiple sensors for detection of events and actions that are confirmed by one or more other sensors enables the generation of a reliable data set due to increased accuracy of one or more reliability and / or probability metrics of that data.

[0088]

[0097] Environmental Sensing The relationship of a PUM to its dominant residential environment can form the basis of a comprehensive set of reliable behavioral patterns for that PUM and the environment. A key aspect of the system is to establish the state of the PUM in the environment such that the person's behavior within at least one particular pattern of behavior can be determined with high accuracy. To achieve this degree of accuracy, the environment can be sensorized to generate a data set that establishes a context for these behavioral patterns.

[0089]

[0098] One or more classes of sensors capable of active emission technologies, such as RADAR, Light Detection and Ranging (LIDAR), Radio-Frequency (RF) and / or audio and / or other emission-based sensing technologies, can be used to create a representation of the environment being monitored. These emissions can be combined to create a 3D map of the environment, including any fixed features within that environment. The mapping can be augmented and / or built upon by 2D and / or 3D drawings of the environment. In some embodiments, these drawings can form part of the environmental framework, which can then be populated with active emission data. The result of this process can be a 3D model of the environment, which can then be converted into a digital twin of that environment.

[0090]

[0099] In some embodiments, active mapping techniques can be deployed using, for example, a frequency modulated continuous wave (FMCW) radar to create a radar phase map of the environment. This can incorporate multiple modulation patterns, such as sawtooth, triangular, square, sinusoidal, and stepped. Each of these can result in different characteristics of the environment under evaluation. The use of different types of modulation can improve the identification of the boundaries of the environment and the shapes of objects within the environment. Each type of surface, such as glass, wood, plasterboard (drywall), and other hard surfaces, produces specific characteristics. The same is true for soft surfaces, such as carpets, curtains, couches, etc. Given the capabilities of FMCW, this can result in an accurate representation of the environment. The use of multiple modulations provides further refinement of these representations. Additionally, these data sets may be passed to one or more care analytics management processors for classification and / or identification. In this manner, a repository of object types, wall and floor types, windows, benches, fixtures, etc. can be developed for use in further representations of other environments.

[0091]

[0100] Such approaches may include the use of algorithmic techniques that may be deployed, for example, interferometry, which uses relative interference patterns produced, for example, by returns from stepped, multiple frequency radiation, to identify particular objects (e.g., furniture) and surfaces (e.g., windows) present within the environment.

[0092]

[0101] The combination of emission-based active mapping can include, for example, the use of stepped frequency radar mapping, where the use of phase and multiple frequencies can identify environmental fixture resonances that can be used to reduce ambiguity. For example, a single pulse (or a simple set thereof) can be used as an active sweep when such a sensor is activated to identify or confirm the location of an object or person within the environment.

[0093]

[0102] A typical environment, such as a home or care environment, has a set of built-in devices that can act as sensors, which can include "night light" motion detectors, door entry / exit triggers, window open / close triggers, smart light bulbs, smart temperature / environment controllers, smart TVs, smart speakers, smartphones, other Internet of Things (IoT) devices including smart appliances such as refrigerators. This can include any device worn and / or carried by the person under supervision and / or other parties. These can be integrated into the care and health system to the extent supported by their characteristics and / or configuration. Typically, such integration is via one or more APIs and / or standardized messaging systems.

[0094]

[0103] An environment may be profiled using acoustic emitters and receivers to determine a baseline acoustic state of the environment. This can then be used by at least one machine learning system to establish an acoustic quiescent state of the environment. This baseline can then be deployed to identify, in whole or in part, acoustic signals in the environment, such as when the PUM is walking, sitting, standing, opening and closing doors, opening and closing drawer windows, or performing other actions. The use of the background acoustic quiescent state can assist in the recognition and identification of other acoustic events, which can potentially be stored as fingerprints on the edge device, allowing matching and comparison techniques to be deployed.

[0095]

[0104] Another class of sensors using passive sensing such as computer vision includes cameras, including the use of sensors with IR or other non-visible spectrum capabilities, which can be used to map the environment, which can include the use of panoramic capabilities such as on smartphones. Often the contents of the environment are private to the person inhabiting it, and therefore the images produced in such mappings can be encrypted and any identifying features removed, with the functional aspects being retained in the digital twin, for example as an avatar or other representation. There are many identifying features of the environment that are not required for the effective functioning of the system, and therefore these can be represented in generic forms (e.g., books, paintings, objects, etc.) that are represented as their functional equivalents in terms of mapping the environment. For example, a book represented as a soft surface, or a book represented as an acoustic signature.

[0096]

[0105] In some embodiments, one or more sensors can be used to create a fingerprint of one or more wavelengths, such as RF, IR, visual, acoustic, etc. This set of fingerprints can include one or more sensor data outputs, such as thermal and visual or RF and acoustic, and can form part of a pattern of the environment, e.g., the quiescent state of the environment. This support for detection and identification of variations in this state (e.g., changes in the pitch of an air conditioner or additional heat generation from an appliance) can indicate a potential care or health issue for the involved person. Such techniques can also be used for preventive maintenance, for example, where loss of functionality of an appliance, such as an air conditioner in a high heat zone, can cause a significant health or wellness event.

[0097]

[0106] In some embodiments, one or more cameras can be deployed to capture images of the environment and its contents. These cameras can include capture of both visual and non-visual wavelengths. These cameras can be part of other devices, such as smartphones, smart TVs, smart speakers, security or other systems, and / or can be specifically deployed to monitor the environment.

[0098]

[0107] In some embodiments, each camera may have access to some processing capability, e.g., this may be part of the camera and / or the host device and / or may be accessible via one or more communication capabilities. For example, a camera may have sufficient local processing to detect changes in a surveillance image, e.g., image processing, and then invoke one or more systems that provide further image processing, such as image recognition and / or other data extraction, comparison, matching, etc. functions.

[0099]

[0108] In some embodiments, cameras and other image capture sensors and devices can include one or more machine learning capabilities that provide image recognition, including video and motion, so that various conditions, including the environment and one or more participants, can be detected, identified, measured, and / or classified. This can include obfuscating an image or set of images so that certain distinguishing features can be recognized without fully disclosing the details of the image. In this manner, the privacy of the PUM or other participants can be maintained while data specifically related to actions and / or events that impact the care and well-being of the PUM is identified.

[0100]

[0109] In some embodiments, a camera may be used to verify an event, such as a PUM lying on a couch, and other sensors, such as a smartphone sensor, may interpret it as a change in orientation associated with a fall. In this example, the camera may provide data that provides context for the event, thereby reducing false positives. In another example, image processing of the camera and any connected systems may identify the event as a pattern consistent with a fall, thereby verifying a health event. In some embodiments, an event, such as a fall, may be recorded by the camera and kept for access by one or more authorized parties, such as a paramedic, and this record may provide important data regarding how the PUM or other parties fell, as well as the potential medical and / or health effects of the fall. In other situations, the data may not be recorded, but edge recognition may be used to keep the trajectory of the fall for further pattern recognition. In some embodiments, the fall may be recorded to allow one or more systems to substantiate the occurrence of the event and its circumstances, for example to verify an insurance claim, etc.

[0101]

[0110] In some cases, some of the fixtures and / or contents of the environment may be moved, removed, and / or added to the environment. The separation of fixed and mutable fixtures of an environment is made in both the environment framework specification and in any digital twin representation of that environment.

[0102]

[0111] For objects important to the monitored person, such as those with memory, vision, hearing, mobility or other impairments, the system can invoke one or more active sensing technologies, including radiation, beacons, cameras, tactile, etc., to maintain up-to-date locations and precise positioning of those objects deemed essential to the person's well-being. This can also include convenience functions for the PUM, such as finding misplaced glasses, smartphones, books or other items.

[0103]

[0112] 4 illustrates a PUM having a set of attached wearable devices, a set of dedicated care sensors, and a set of general-purpose sensors, all of which are located within the environment in which the PUM resides. These sensors may be configured as edge sensors and / or may be in communication with one or more care analytics management processors.

[0104]

[0113] The integration of active and passive sensor data can represent a state from which any variations can be detected, and such representation of the state is updated if and / or when necessary. For example, moving furniture, installing new equipment, etc. In some embodiments, the digital twin can be used as a representation of the state of the environment, thus providing a history of the environment and its contents. Such history can then be used by one or more machine learning and / or pattern recognition systems to establish causal or correlational relationships between care and health events and the contents and configurations, including positioning, use, avoidance, etc., within the environment.

[0105]

[0114] Exemplary embodiments Mobile personal emergency response system (PERS) devices are generally targeted at elderly and / or disabled or other persons who need to request help or emergency services, for example, by pressing an emergency button on the PERS device. These devices typically include an emergency button, a speaker, a microphone, and wireless communication capabilities, including, for example, wireless telephone capabilities used to connect a person with emergency personnel or caregivers using voice. In some cases, PERS devices may also include sensors and software that detect events, such as a fall, and use the sensor's signals to automatically trigger an emergency call and / or report the event to a central server and / or service when such an event occurs. They may also include location detection sensors, such as GPS, radio frequency triangulation, beacon readers, etc., that enable the PERS device, or the system to which it connects, to trigger an emergency and / or other response, for example, when a person leaves a predefined area (geofence), stops moving for a sufficient amount of time or other location and / or movement-related circumstances.

[0106]

[0115] One problem with a PERS device configured with these and other sensors is that keeping all sensors active most of the time and processing their signals to effectively detect relevant events can very quickly drain the device's battery, thereby reducing its usefulness in real-world situations. By applying the embodiments described herein, this problem can be addressed and the performance and accuracy of the PERS device's functions can be improved. For example, in normal situations (such as a stationary state), most of the sensors in the PERS device can be configured to remain stationary except for edge sensors, such as an accelerometer, so that software in the PERS device can only listen for signals from the accelerometer that indicate movement above a predetermined threshold, indicating that the person has changed state from a set of activities consistent with a normal pattern to one or more activities that exceed one or more thresholds of that normal pattern. At this point, other sensors, such as an altimeter or other height detection sensor, and / or a microphone and / or other sensors in the device can be activated. In this example, the configuration of the edge sensors, such as the accelerometer and thresholds for its data, can be changed so that the system and software can switch to a different set of detection logic to create a different configuration suitable for detecting the most likely event under the person's new state. Additionally, other biometric sensors may be activated, as well as location detection sensors and / or geofence logic. In some embodiments, these configuration changes may be dynamically provided to one or more sensors, either proactively or in response to a change in condition detected by one or more sensors, including edge sensors.

[0107]

[0116] For example, a geofence can also trigger a new configuration change, e.g., when "going out" of the environment is detected, and the operational parameters and / or event detection logic of sensors such as accelerometer and altimeter can be changed to detect the dynamics of walking outside, or stopped if a "moving in car" situation is detected, e.g., based on a combination of position change and accelerometer data. In this approach, sensor, processing, and communication capabilities are used only when detected patterns indicate they are needed, resulting in reduced power consumption. Additionally, the dynamically changing sensor configuration and detection logic allows for improved event detection accuracy.

[0108]

[0117] An implication of using a PERS device as a single method of detecting risk-related events, such as falls, is the accuracy limitations that result from a set of co-located sensors in a small, portable device. This makes it difficult to avoid false positive and false negative event situations. This can be improved by combining the PERS device with the sensor configuration, data processing, and detection logic of a device in the same environment but external to the PERS device, as described herein.

[0109]

[0118] For example, a PERS user's home environment may be equipped with additional sensors, such as cameras, smoke detectors and / or microphones, as well as a range of other sensors. Signals from these sensors may be combined with data from the PERS device's sensors, as well as data from other devices, such as voice recognition enabled speakers (such as smart speakers and / or smart TVs), as a way to expand and extend the capabilities of the PERS system to increase its effectiveness. This combination may occur in one or more systems, including those hosted within the PERS device and / or other devices in the user's home, and / or on one or more remote servers connected to the PERS device using wireless communication mechanisms such as Wi-Fi and / or Bluetooth. This combination may be used to more accurately determine the user's condition based on known and / or expected user behavior patterns, such as typical locations and activities within the home, and the signal patterns those activities generate at the sensor placement of environmental sensors in combination with the sensors of the PERS device.

[0110]

[0119] Some of the devices and / or systems and / or system and / or server configurations in this type of configuration may include machine learning and / or statistical mechanisms as an adaptation method to identify patterns indicative of changes in the user or environmental conditions, to more accurately and / or dynamically select sensor configurations, and / or to trigger events, alarms and / or responses. Signals transmitted by sensors in the PERS devices and the user's environment may be stored and used by one or more servers to train machine learning systems and / or to provide Care Village Digital Twins (CVDTs) that represent the user and its environment, supporting adaptation to changes in the user's behavior and / or environment. This approach helps to increase the accuracy of predictions to better predict and prepare emergency and / or other assistance resources in response to changes in detected patterns.

[0111]

[0120] FIG. 5 illustrates a PUM (101) residential environment including one or more sets of sensors including wearable, portable and / or mounted sensors / devices, and / or sensors located within the environment including dedicated devices and / or sensors and / or general devices including sensors.

[0112]

[0121] In some embodiments, a "care VPN" can be instantiated, allowing designated authorized and authenticated parties to access, control, configure and / or monitor one or more sensors, devices and / or systems in the environment. For example, a dedicated device providing functionality to sensors in the environment may store a volume for data prior to a care and health incident and make it available to an EMT, paramedic, caregiver, or other authorized party when in proximity to the PUM and the dedicated device detects, determines, presents, or activates. For example, this may be the case when the PUM is experiencing a care incident and makes an expert on that type of incident aware of the situation, for example.

[0113]

[0122] In some embodiments, a "caregiving VPN" is instantiated between a specialized device and one or more parties, enabling data managed by such a device (including data stored by the specialized device) to be made available to one or more authorized and authenticated parties. In some embodiments, this may include the use of secure tokens for such access.

[0114]

[0123] The previous description of the embodiments is provided to enable any person skilled in the art to practice the present disclosure. Thus, the present disclosure is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0115]

[0124] Sensor Type In some embodiments, there may be one or more specialized sensors designed to detect specific events, including those that are particularly related to the particular care and health of one or more persons in one or more environments and / or that provide an indication of fluctuations in care status. These sensors may individually or collectively in any configuration provide data to one or more systems to establish one or more patterns of the environment and / or the participants inhabiting the environment itself.

[0116]

[0125] In some embodiments, a set of sensors can be integrated and deployed within an environment in a manner designed to ensure that the behavior of participants, including the PUM, is effectively monitored in support of the care and well-being of that PUM. Each environment can incorporate a different arrangement of sensors to provide sufficient coverage of the environment to establish and monitor patterns of behavior therein. In some embodiments, various combinations of sensors, including those worn or carried by the PUM, are deployed to establish conditions of the environment and / or its participants that support the pattern or patterns being monitored. Some examples of such sensors and / or devices are described herein.

[0117]

[0126] Strain Gauges

[0118]

[0127] Strain gauges can be used to detect the range of variation in the degree of strain experienced by the surface to which they are attached. For example, if a strain gauge is attached to a floor and there is an impact from a falling object, a pet or a person, the strain gauge can detect and measure such an impact.

[0119]

[0128] Strain gauge sensors can be wired and / or wirelessly connected to one or more systems that monitor their operation. The sensitivity of the strain gauges can be altered by electronic and / or mechanical methods and / or one or more signal processing and / or data processing systems. In many situations, the strain gauges have electrical connections that allow for measurement of the impact detected by the strain gauge. Temperature can affect the sensitivity of the gauges, so this can also be measured and used to construct measurements from the strain gauges.

[0120]

[0129] One aspect of this approach is the measurement of various impacts, both intentional and unintentional, such as those caused by footfalls of PUMs, or other stakeholders walking in the environment, or by falls, trips, or other oversights. Strain gauges can also detect and measure the impacts of pets, including, for example, a cat landing on a surface, or a dog lying down.

[0121]

[0130] The strain gauges can be deployed such that a measurement of an impact can be detected by each gauge, differing with respect to the location of the impact within the environment, for example, positioned in a corner of a room. When the strain gauges are connected to one or more systems, these multiple data sets can be combined to allow evaluation of the data, for example, to determine the speed, acceleration, mass, direction, location and / or other characteristics of the impact. This can provide, for example, a PUM walking through the environment, a pet crossing a room, and / or a pattern of impacts indicative of a care or health event. In some embodiments, such data generated by one or more strain gauges may be provided to one or more machine learning, pattern recognition, matching, or other analysis and / or evaluation systems to determine, in whole or in part, patterns of movement and / or other impact occurrences within the environment.

[0122]

[0131] Strain gauges may be attached to or integrated into the surface they are monitoring. For example, strain gauges may be attached to a wooden floor or other hard surface that is highly rigid, and / or woven into a softer material such as carpet. In the carpet example, the pattern of strain gauges may be such as to form one large gauge, several independent gauges, or a set of gauges in any arrangement.

[0123]

[0132] In some embodiments, the strain gauge or set of strain gauges may form part of a sensor. For example, the sensor may include communication, processing, memory, and / or other elements. For example, a sensor including one or more strain gauges may have sufficient processing power to determine an impact or set of impacts that can be pattern matched to a previously identified impact or set of impacts such that accurate identification of the pattern is achieved. This identification information may then be transmitted, for example in the form of a token, to another system, sensor, and / or other authorized entity. This may include a sensor incorporating a set of tokens that represent different occurrences, including conditions of the environment such as quiescent conditions. The token may include one or more metrics, such as the degree to which the occurrence matches the pattern, which may be expressed as a confidence level, for example.

[0124]

[0133] In some embodiments, a set of strain gauges can be used as a network, and machine learning can be used to evaluate input signals, some of which may be of low intensity, and one or more signals can be detected, for example, using neural networks or other undirected learning techniques.

[0125]

[0134] In some embodiments, the strain gauges can be designed in the form of antennas to match any of the typical Wi-Fi and / or Bluetooth frequencies, such as 2.4Ghz, 5Ghz, or 6Ghz. Each of these deployed strain gauges can provide a signal to a central server when the antenna is energized by absorbing transmitted RF energy. For example, if the output of the strain gauge is within a range determined as the resting state of the strain gauge, this does not represent measurable activity. For example, FMCW radar can be used to energize the strain gauges, and these strain gauges are formatted to match the antenna of the FMCW radar signal transmitted from, for example, a smart light bulb. In both of these examples, this eliminates the need for the strain gauges to be wired when energized by RF, and thus a monitoring system operating at the same frequency can detect the variation in the strain gauge output when experiencing a change in measurement caused by one or more impacts.

[0126]

[0135] If an event occurs, such as a fall, that interrupts the signal from the strain gauges, this data is passed to one or more central servers and / or edge devices containing the sensors and used to trigger one or more events, including responses, alerts, and / or sequences thereof.

[0127]

[0136] Patterns of movement typical of a person in the environment are also captured by the strain gauges, in that a person's stepping also generates a resting pattern, which, although smaller in magnitude than a fall, can still be determined using, for example, machine learning techniques. Such movements form patterns that can then be incorporated into behaviors or other broader activity patterns for the PUM. For example, if a PUM regularly traverses the environment from the bed to the bathroom, correlating this with time can indicate use of these facilities.

[0128]

[0137] For example, the learning phase for machine learning is the recording of a steady state of strain gauges with at least one transmitting source of an appropriate frequency, where no person is present. In this way, a baseline of the environment is established. The system is then set to a second learning phase, where a person uses the environment and his / her footsteps are recorded as a further "natural" state of the relationship between the person and the environment. This can then form part of the stationary state of the environment, where PUMs are present and performing their natural and normal activities, from which any deviations can be detected.

[0129]

[0138] By using machine learning to establish baseline patterns of relationships between the environment and the person, normal behavioral characteristics can be established against which events and alerts, such as falls, can be determined.

[0130]

[0139] Strain gauges can be attached directly to any surface, primarily the floor of the environment for detecting falls. This connection can be direct or indirect, for example, stain gauges can be attached to furniture or other objects and devices that are connected to the surface of interest. This can include both hard and soft surfaces.

[0131]

[0140] Windows and Surfaces

[0132]

[0141] The habitable environment has one or more windows with at least one surface exposed to the interior of the environment that may be used for some sensor placement and may incorporate one or more sensor capabilities.

[0133]

[0142] For example, through the attachment of a suitable sensor, such as a MEMS microphone or other suitable acoustic sensing device, a window can be used as a microphone. In this way, a window in an environment can capture acoustic information in the environment.

[0134]

[0143] This detection may include the use of speech recognition techniques, determined by the identification of key phrases, such as, for example, "help," "fall," "assist," or other terms that a party, such as the PUM, may provide to the system during a learning phase. This learning phase may be configured, for example, such that the PUM moves around the environment repeating certain words or phrases designated as trigger phrases that may be detected by one or more sensors, including any device incorporating a microphone, and recognized by one or more systems, which may then invoke one or more responses.

[0135]

[0144] Such acoustic capture sensors, such as, for example, MEMS microphones or other microphones, such as those included in smart speakers, smart TVs, smart watches, smartphones, PERS and / or other sensors, devices and / or systems, can be used to detect and identify intonations in one or more speech patterns of one or more participants. For example, this can include detection of stress, anger, frustration and / or other speech related features, including repetition or specific phrases, etc. Such techniques can also detect other sounds, such as, for example, sounds made by a participant when experiencing a fall or other health and / or care related event. This can include sounds representative of experiencing more or less pain or discomfort, such as, for example, grunting, yelling, shouting, leaning over, etc.

[0136]

[0145] These and any other microphones may have sets of filters applied, for example, to attenuate or emphasize sounds, and may have high, low or band pass filters applied to detect sounds representative of and / or associated with particular occurrences, e.g., falls, collisions, breaking of furniture or objects, etc. In some embodiments, specific filter sets, e.g., using a DSP, may be applied, for example, at a stationary monitoring stage or when an event or trigger is detected.

[0137]

[0146] This can include beamforming to directionally focus microphones directly and indirectly on detected sound sources of interest. This can be invoked, for example, by another sensor, where the first sensor detects a change in condition of sufficient magnitude, e.g., an increase in acceleration, a change in height, or other attribute detected by the worn device, such as PERS, and sends a message directly or via or in conjunction with a service to the second sensor for confirmation of the detected event.

[0138]

[0147] The filtering applied to the microphones may be dynamically changed in response to an event, either directly or via another device, to focus on frequency ranges, amplitudes or other characteristics associated with voice or other speech from a person, particularly in association with particular words or phrases previously stored by one or more systems and / or in response to detection of sounds indicative of an impact, fall or other occurrence affecting the care or health of the PUM.

[0139]

[0148] The use of other flat surfaces for mounting microphones, such as MEMS microphones, can also be used. Windows are particularly well suited for capturing acoustic signals given the properties of glass, but other materials such as plastic, wood, etc. may be used. Many environments have an expanse of flat surfaces that can be used to capture acoustic data. Often, larger surface areas can be used to capture longer wavelengths, for example those of low frequency acoustic events such as those generated when a PUM or other party falls or strikes a surface.

[0140]

[0149] For example, MEMS microphones may be deployed on hard surfaces, potentially in combination with strain gauges as detectors of shock. In some embodiments, these MEMS microphones can be attached to hard surfaces such as walls, windows, floors, etc., and thus can detect acoustic vibrations and disturbances that represent changes in stationary or other conditions of the environment.

[0141]

[0150] The use of MEMS microphones or other microphones in the environment can be used to track acoustic events both directly and indirectly. For example, a MEMS microphone or other acoustic sensor on or near bathroom plumbing can indicate the use of bathroom features without interfering with the privacy of the person using the bathroom.

[0142]

[0151] The relative positioning of these microphones and / or other acoustic sensors can provide different signals which, when processed with appropriate digital signal processing techniques such as Fast Fourier Transform (FFT), phase shift analysis, filtering, digital signal processing (DSP), etc., can provide data sets that can inform regarding activity in the environment separately and / or in other arrangements, such as the location of people in a multi-room environment, dwell time in specific locations such as bedrooms, bathrooms, kitchens, etc. as people move through the environment, etc.

[0143]

[0152] Acoustic fingerprinting of an environment, in part or in whole, may be accomplished using suitable acoustic transmitters and such MEMS or other types of microphones in any arrangement, however in most situations there will be locations in the environment that are favorable for the location of such devices, for example hard surfaces such as ceilings, walls of hallways, etc.

[0144]

[0153] The use of a set of MEMS or other microphones in an environment can be configured such that the microphones provide sufficient coverage of the environment to generate an acoustic fingerprint of that environment, in whole or in part. This fingerprint contributes to the state of the environment and can be part of the environment specification, including part of the static state of the environment. The output of each microphone can be evaluated individually and / or collectively in any arrangement to determine one or more patterns of activity, movement, and / or behavior of a participant within the environment. This can include establishing acoustic fingerprints for various activities and actions, such as traversing a hallway or room, etc., such that a state of activity of a participant within the environment can be determined generally in relation to one or more patterns.

[0145]

[0154] In some embodiments, the arrangement may be such that the microphones are positioned to create geometric relationships, such as on opposing walls, forming a triangle or other shape. These geometric relationships may then be used, for example, by the care analysis management processor as part of the signal analysis. For example, phase, amplitude, reflection, time and / or frequency relationships may inform the care analysis management processor's evaluation.

[0146]

[0155] Heat Detection

[0147]

[0156] In some embodiments, environmental sensing can include the use of thermal sensors to establish sources of thermal radiation within an environment, which can include an initial survey in which no PUMs or other participants are present, followed by forming a baseline for further sensing by such thermal sensors to establish the location of the PUM in such environment.

[0148]

[0157] Thermal imagery and / or temperature derived data can be used to establish the presence of a person in a portion of the environment and / or to establish the condition of the person and the environment itself. Infrared sensors can be used to determine relative different temperatures within the environment and, in combination with visual cameras, can determine the cause of such temperature differences. For example, in inclement weather, thermal sensors can provide data indicating that temperatures within the environment are likely to cause or affect a care or health event. For example, if the external temperature is excessively low or high, for example, during a winter storm or summer heat wave, the temperature of the environment may cause the PUM to be exposed to excessive cold or heat, respectively. This thermal sensor data can be used to operate an HVAC or other climate control system, but in some circumstances, these systems may not be able to provide sufficient mitigation of the external condition. In this situation, one or more alerts or events can be generated by the care village system, including, for example, alerting a caregiver, neighbor, friend and / or family member of the situation. This can also include providing a message to the PUM, for example via a smart speaker, to change clothing or otherwise mitigate the effects of the external condition. This can be particularly important with respect to memory impairment.

[0149]

[0158] Uses of heat detection can include identifying environmental activities such as use of cooking appliances, running showers or baths, heating and cooling, etc. For example, such heat sensor data may be used in conjunction with other sensor data to verify and validate that PUM behavior patterns are consistent or that there are changes from those patterns.

[0150]

[0159] The use of thermal sensors can also help determine the movement of PUMs, including, for example, their presence in bed, traversal between rooms in the environment, etc.

[0151]

[0160] The thermal sensor can also provide data indicative of the overall temperature of the environment, such as the temperature during a weather event. For example, if there is a high external temperature, the sensor data can indicate that the environmental internal temperature is approaching a threshold that could cause the PUM to experience a health event, such as heat exhaustion.

[0152]

[0161] Motion detector

[0153]

[0162] Many environments have motion detectors and / or incorporate devices that include such functionality. For example, most home alarm systems incorporate one or more motion sensors. The integration of motion detection functionality into a sensor set can act as an edge device to trigger other sensors in the environment. For example, motion detection can trigger cameras and / or other devices, such as activating a smart speaker, and / or emitting a message or query such as "Hello," "Are you OK," a beep or other audible announcement.

[0154]

[0163] Motion detectors may also be used in various locations to signal the presence of a PUM or a person involved, and possibly the presence of a pet, robot, or other non-human, to one or more sensors, devices, or systems. This can be used, for example, to determine whether a PUM has entered or exited a room or other space.

[0155]

[0164] In some embodiments, the motion detector may form part of a set of sensors and function as an edge trigger device that alerts other sensors, devices, and / or systems to the presence of a PUM or other party. This is particularly useful where these other devices have power requirements that require them to conserve their resources so that they are only active when triggered. This may also be true if the motion detector no longer detects motion.

[0156]

[0165] In some embodiments, the motion detector can be used to change the configuration of one or more sensors, devices, and / or systems. This can include configuration to enhance or extend sensor functionality, such as turning on a camera or activating an active sensor. Such configuration can also be used to reduce or limit the operation of sensors, devices, and / or systems, for example, in support of PUM and / or other party privacy.

[0157]

[0166] In some embodiments, motion capture can be detected from one or more image capture systems, e.g., cameras, which have sufficient image recognition capabilities to identify changes in the images being monitored.

[0158]

[0167] The motion detection patterns may form part of one or more behavioral patterns and may indicate deviations from the patterns by the PUM or other parties. For example, if the motion detector indicates that the PUM now has a different set of movements, such as repeatedly using an implement in a short period of time, this may indicate a change in the state of the PUM's short-term memory, indicating further memory impairment.

[0159]

[0168] For example, a motion detector, in combination with a strain gauge and / or PERS or other wearable or portable device, can identify changes in the movement pattern of the PUM, e.g., a hip, knee or other physical aspect is deteriorating, causing a change in their mobility. In some embodiments, the motion. detection can act to configure the strain gauge, PERS and / or other sensors, devices and / or systems to capture and / or measure the movement of the PUM. This movement can then be evaluated and determined to be within a threshold of a movement pattern or to have changed sufficiently from that pattern to indicate a change in the PUM mobility.

[0160]

[0169] Smart Light Bulbs

[0161]

[0170] The emergence and availability of smart light bulbs offers another sensing capability, in that such devices incorporate radar technology that can determine certain biometric functions such as heart rate, body temperature, etc. In some embodiments, such devices can be integrated into an environment to provide ambient lighting that can be integrated into one or more environmental sensing systems to provide care and health benefits to one or more parties occupying the environment.

[0162]

[0171] One technology employed in such bulbs is the use of FMCW (frequency modulated continuous wave) radar in the 2.4Ghz range. In some embodiments, a set of smart bulbs can include an FMCW mesh. For example, if each room in a multi-room environment has one or more smart bulbs, the biometric data of the PUM can be tracked seamlessly from room to room.

[0163]

[0172] One aspect of the use of smart light bulbs is their ability to non-invasively track sleep patterns. Data from such devices can then be integrated into one or more patterns to provide a comprehensive data set to assist stakeholders in assessing health conditions. This is especially true when such data is integrated into one or more digital twins, which can then be used to inform one or more predictive systems for the identification of changes in behavioral patterns that indicate potential health or care events.

[0164]

[0173] In some embodiments, FMCW radar can be used in devices specifically designed to track the care and health data of participants. Such devices can be installed within the environment and configured to detect and identify the breathing and / or other care and health patterns of PUMs.

[0165]

[0174] Touch

[0166]

[0175] In some embodiments, tactile detectors may be used to provide the PUM with a method of communication with the system and / or interested parties and / or as a sensing method for determining the operation of the PUM. For example, surfaces that are prone to being grasped as a person tries to regain their footing or avoid a fall, such as banisters, may be provided with tactile detectors that can detect such activity.

[0167]

[0176] Haptics may also be worn as an addition to and / or as part of a garment, both for monitoring and providing communication. For example, a small pad may be attached to or incorporated into the garment such that applying a touch to the pad can create one or more messages for communication to one or more sensors, devices, and / or systems. For example, different touches of a specific one finger, multiple fingers, thumb, palm, etc., any combination thereof, can generate different messages. This may include, for example, an emergency call to 911 or other emergency services following a fall or breakdown, where the person involved does not have access to another device but can touch the pad, providing a haptic signal.

[0168]

[0177] In some embodiments, the pad can be worn on the forearm, e.g., near the wrist, and can include a power source, including one that generates energy from motion, one or more communication capabilities, and one or more identifiers, which can include specific serialized indicia and / or antennas that can be recognized by one or more sensors, devices and / or systems, such as QR codes or other types of bar codes, electronic identifiers, etc.

[0169]

[0178] This allows the wearer's movements to be tracked, particularly with respect to entering and leaving an environment or areas within that environment. Such pads may include one or more biometric sensors and / or alignment indicia, any of which may be used for detection by one or more sensors, devices and / or systems of a person changing alignment within the environment. For example, when the PUM moves from a standing position to a sitting or lying position.

[0170]

[0179] In some embodiments, tactile sensors may be mounted in the environment to provide communication capabilities to PUMs or other parties that do not have another way to communicate their situation. For example, in a bus, a pad may be on the side of the tub or easily reachable from within the tub, and if the PUM has a care or health event, the pad can use the tactile pad to communicate with one or more sensors, devices, and / or systems to call for assistance. Other potential locations may include stair cases, areas adjacent to the floor, areas where falls are likely to occur, etc.

[0171]

[0180] In some embodiments, the tactile surface or pad may be connected to one or more other sensors, devices and / or systems. Such a surface and / or pad may be configured to function as a mouse pad for controlling, for example, one or more computing devices, appliances such as smart TVs, audio systems, HVAC, stoves and ovens or other appliances. In some embodiments, this may include a haptic interface device that receives inputs from the haptic device and then translates these communications into commands and configurations for one or more target devices and / or appliances. This may include communications that create command and configuration sets for multiple devices and / or appliances, for example simultaneously or in a time-ordered arrangement.

[0172]

[0181] In another example, the tactile surface may be embodied in a yoga mat, or other functional surface including, for example, a bed coverlet, a shower mat, a bath mat, etc. In this example, the tactile surface may function as both a sensor and a communication system, with different areas of the surface taking on either or both of these functions. This may include configuration of the surface from one function to another, either directly or remotely. For example, a PUM may place their hand on a tactile pad, which then configures the bath for their particular needs.

[0173]

[0182] In some embodiments, tactile sensors can be deployed such that they can create a data set of the strength of the PUM used as it grips a surface, such as a door handle, stair rail, refrigerator door, etc. This data can then form part of the PUM's behavioral pattern and potentially indicate variations in the strength applied, which can indicate care and / or health issues. This can also include walkers, ziggurat frames, canes or other mobility assistance devices.

[0174]

[0183] The integration of a series of tactile sensors and / or control surfaces within an environment can provide a data set that can be integrated into one or more behavioral patterns.

[0175]

[0184] In some embodiments, where the use of cameras is considered invasive, a CCD type detector may be used and edge determination algorithms may distinguish the shape of the PUM to the extent necessary to establish their orientation, e.g., standing, sitting, lying relative to the environment. In this manner, such sensors may be deployed in restrooms and the like where many falls occur without compromising the privacy of the PUM. In some embodiments, such CCDs may capture full resolution video that is stored over the period that the PUM is using the bathroom or other facility, but this data is only available to one or more authorized parties in the event of an emergency, such as a fall, where such data is determined to be essential to the care of the PUM. For example, if the PUM falls and hits his head on the bathtub, such data may be essential to the care of the PUM. Even in this example, there may be blurring or other privacy features that can still provide appropriate data while protecting the privacy of the PUM.

[0176]

[0185] Wearable, portable and mounted devices

[0177]

[0186] There is a wide range of devices, including various types of sensors, that the PUM can carry, wear, or attach to itself. Certain wearable devices may also be attached to other participants and / or pets residing in the same environment. These can include devices such as personal emergency response devices intended to identify critical care conditions and alert designated participants to the occurrence of such events. There are also numerous devices that can be integrated into the sensor-enabled environment to provide a data set that can be integrated into one or more patterns for the environment, the PUM, and / or other participants, such as smartphones, smart watches, fitness trackers, smart wallets, biometric devices (including biometric headbands, EEG / ECG sensors, etc.), sensor-enabled clothing, smart glasses, smart footwear, smart jewelry, smart earphones, etc.

[0178]

[0187] The trend towards integrating electronics, sensors, and more recently haptics, into wearable devices allows for the integration of multiple sensors into care village systems. These range from those with specialized functions, such as blood pressure monitors, to those with a set of sensors, such as fitness trackers, smart watches, etc. This includes sensors embedded in clothing or other wearable materials.

[0179]

[0188] Many of these devices may be designated as edge devices in that they provide an initial set of data that activates, alerts, and / or triggers other sensors, devices, and / or systems. While the proximity of these sensors and devices to the PUM ensures that the data they provide is accurate, there may be situations where the context of the data reveals a more accurate perspective on an occurrence. This is particularly the case for false positives. For example, if the device detects a sudden change in direction, speed, and acceleration, as detected by a smart watch, for example, this may be determined to be a fall, when in fact the PUM has tripped and turned on its own. Triggering other sensors, for example cameras, and / or non-triggering additional sensors, for example strain gauges to detect impacts, can provide context of the occurrence and reduce false positives.

[0180]

[0189] The integration of sensors and devices into wearable clothing or materials is expected to continue, and thus the care village system will integrate these sensors and devices into the environmental sensing system. This integration may include receiving data sets from these sensors and devices, and configuring such devices in light of the situation. For example, they may be turned off during certain activities, their sensitivity may be increased or decreased, their cached data sets may be expanded, and short recorded periods may be cleared.

[0181]

[0190] In some embodiments, the PUM may have one or more movement aids, such as a cane, a Zigmar frame, etc. These movement aids may be fitted with one or more sensors, including haptics, which may be integrated into environmental sensing. These sensors may contribute to one or more patterns, and may also be configured as edge sensors to detect care and health events. In some embodiments, these movement aids may incorporate other devices to mitigate falls or other health events. In some embodiments, these may be considered smart mobility devices, and the potential range of sensors may include, for example, motion detection (accelerometer, gyroscope, altitude / barometric pressure), haptics, impact detectors, cameras, radar / lidar, etc. There may be some of these devices equipped to be attached to and / or integrated into the mobility device, as well as other mobile devices, such as PERS, smartphones, etc.

[0182]

[0191] In some embodiments, the shoe or other footwear may include one or more sensors that may communicate with another wearable device, sensor, device, and / or system that is part of, for example, environmental sensing. For example, a pressure sensor may measure the pressure of the PUM's foot on or inserted into the shoe to establish a baseline of the person's normal mobility. This data may form part of the PUM's pattern, and if such measurements fall outside of the pattern, the sensor may alert one or more other sensors, devices, and / or systems. For example, if the PUM trips but does not fall, these missteps may be evaluated to establish any patterns, e.g., increased missteps that may predict an increased likelihood of a fall. This data may also be used to evaluate the environment in terms of removing obstacles that may result in or increase the likelihood of a fall, thus benefiting the care and health of the PUM.

[0183]

[0192] In some embodiments, such footwear may include integration with other smart clothing and wearables, which may include one or more small or micro-airbags that can be deployed when a fall is detected by the footwear and / or other sensors.

[0184]

[0193] Radiation Technology

[0185]

[0194] The range of devices incorporating RADAR / LIDAR and other RF, Visual, and line-of-sight radiation is expanding due in part to the development of autonomous and semi-autonomous devices, including vehicles. These radiation-based sensors can be deployed within a care village environment, for example, to identify and locate participants, other sensors, devices, environmental furniture and fixtures, etc. For example, a PERS or similar device worn and / or attached to clothing can provide forward-looking data regarding, for example, changes in walking trajectory or slope of a surface such as a staircase. In another example, the sensor can provide data regarding the distance of a wall or other boundary.

[0186]

[0195] In some embodiments, IR technology can be deployed to provide night vision, for example via IR-enabled cameras mounted within the environment and a portable device capable of displaying images captured by such cameras. For example, this can be used at night when there is a power loss and participants can then receive images of the environment, for example requesting them to move to an exit or other safe location.

[0187]

[0196] FIG. 4 shows examples of such devices, and FIG. 5 expands on those examples, showing dedicated care sensors, as well as general devices and sensors that may be installed and / or embedded in the PUM's living environment.

[0188]

[0197] Environmental Sensing

[0189]

[0198] In many environments, smoke, CO2, carbon monoxide, and other environmental sensors are deployed for the safety and well-being of participants within such environments. These sensors, as well as other gas, atmospheric, and other environmental sensors, can generate data indicative of the air quality within the environment. In some embodiments, multi-purpose devices such as smartphones, smartwatches, and the like can include such sensors, as can specialized devices intended to monitor atmospheric conditions within an environment. As with other specialized, single and / or multi-function devices, the data generated by these sensors may be incorporated into the care village monitoring system. Such data can form patterns that can be included within the overall monitoring pattern of the environment and one or more participants.

[0190]

[0199] In some embodiments, data from these sensors can be used to trigger other sensors to verify the occurrence of an event and / or contribute to a dataset indicating a high likelihood that a health or care event will occur. These sensors can also provide data indicative of the presence of PUMs and / or other participants in such an environment.

[0191]

[0200] robot

[0192]

[0201] In many situations, there may be one or more robotic devices operating within the environment, such as, for example, a vacuum or other surface cleaner. As the development of autonomous and semi-autonomous devices accelerates, the deployment of robotic devices is likely to increase.

[0193]

[0202] In some embodiments, there may be specialized robotic devices designed to support the care and wellness of the PUM, each of which has a set of integrated sensors that can be configured to interact with sensors, devices, and / or systems in the environment. For example, if the environment has an exterior area, such as a backyard, a drone with an integrated sensor can be deployed when the PUM enters the area. Such a drone can incorporate geofencing, so that it remains within a designated area at a predefined height. The drone can have a fixed charging station attached to the environment.

[0194]

[0203] In some embodiments, the drone may be small enough to be worn, for example, on the PUM's clothing, e.g., on the shoulder, and may be dispatched by the PUM or another party, sensor, device, or system to survey an area the PUM is entering or to provide line-of-sight or other communication capabilities to the PUM. In this example, the clothing worn by the PUM may include charging and docking capabilities for the drone. This may include latches and other connection types.

[0195]

[0204] In some embodiments, there may be a robot designated as a Finder. This device may have mobility capabilities, such as wheeled, tracked and / or limited, that can move through the environment and includes a set of sensors and communication capabilities. The Finder may be used to locate a particular party, most likely a PUM. For example, the Finder may identify the location of a PERS, smartwatch, smartphone or other device worn or carried by the PUM and navigate to that location. The Finder may be equipped with a set of sensors, including sensors that can be accessed by emergency response or other care and are well teamed. For example, the Finder may include cameras, FMCW, microphones, speakers, etc. sufficient for a team to assess the condition of the PUM. For example, in a situation where the sensors, devices, and / or systems are insufficient to assess the condition of the PUM and / or the remote team does not have the communication capabilities to communicate with the PUM, the Finder may provide those capabilities if the PUM falls or damages itself.

[0196]

[0205] The sensors can provide a data set indicative of respiration, heart rate, and other biometric information that can be communicated to one or more authorized and authenticated parties, such as a medical or emergency response team. This can include the ability for that team to speak with the PUM to reassure the PUM and obtain further details from the PUM and / or other parties regarding the PUM's condition.

[0197]

[0206] In some embodiments, the Finder may include some auxiliary functionality, such as being able to lift the head or limbs of the PUM. Such capabilities may be remote controlled.

[0198]

[0207] In some embodiments, there may be a robot that includes sensors to detect floor vibrations, such as one that operates on a floor. This data may then be used to identify and / or detect vibration patterns of one or more participants and can be incorporated into one or more patterns. In some embodiments, there may be room and / or furniture and / or fixture specific sensors, such as pressure sensors placed on, under, and / or in a bed, for example, to detect the presence and / or movement of a participant. A similar approach can be used for couches, sofas, or other furniture intended for sitting, sleeping, or resting. Data sets from such sensors can be incorporated into one or more behavioral patterns of a participant, including a PUM.

[0199]

[0208] Pets

[0200]

[0209] In many environments, PUMs may have one or more pets as companions. Since almost all pets have some form of collar or can be equipped with small portable sensors or devices, these can be incorporated into environmental sensing systems for the care and health of the PUM.

[0201]

[0210] Environmental Conditions In some embodiments, each device with one or more sensors can have one or more states that are determined in part or in whole by the configuration of the device and / or the sensor. For example, if a device including a sensor is generating data, the state of the device and / or sensor can be determined by one or more other sensors, devices and / or systems that configure the device independently and / or in cooperation with other sensors, devices and / or systems. For example, a device that detects a stationary environment, e.g., a room where no people are present and there is no activity in the room, can represent the state of the environment as a stationary state. This can include cases where the sensor is active but the state of the device is stationary. For example, a camera or microphone can receive photons and acoustic vibrations, respectively, but the state of the device is stationary in that none of these inputs represent an occurrence that changes the state of the device.

[0202]

[0211] This approach may include the state of sensors, devices, and / or systems, determined in whole or in part by patterns of data that represent the state of the environment and / or participants therein. For example, if a person is engaging in part of a normal pattern of behavior determined in whole or in part by one or more sensors deployed in the environment, and a data set conforms to a particular pattern that includes thresholds, variations, and / or variances that are part of that pattern, the environment and sensors, devices, and systems may represent a quiescent state of that pattern.

[0203]

[0212] An environment can have many devices within it, each with one or more sensors, and when the environment is unoccupied and no activity is operating, this combination of device states represents the stationary state of the environment, as represented by the aggregation of each sensor state. Such states may change over time, for example, over the course of day / night cycles and over the course of the seasons in which the environment is located.

[0204]

[0213] In some embodiments, external sources of information, such as weather monitoring, can be used to predict and / or verify these changes, which can then be updated with the datasets provided by the devices and their sensing capabilities.

[0205]

[0214] Environments can also have a set of states determined by their purpose, e.g., bathroom, kitchen, laundry, etc. Such specialized environments can have quiescent and specific active states, and people occupying the environments can have dwell times and specific activities related to the purpose of those specialized environments.

[0206]

[0215] For example, these activation states may be specified by either devices and / or sensors that are activated by one or more actions of the system and / or persons within its environment. These activation states may be quantized to form a set of activities having relationships with the environment and one or more persons such that behavioral patterns may be determined, monitored, and / or evaluated.

[0207]

[0216] FIG. 7 illustrates condition management by a care analysis management processor (105) for an environment (102) having a PUM (101), where dates provided by one or more sensors indicate and / or are identified as data representative of condition changes that correlate to one or more behavioral patterns, and each behavioral pattern (107, 127, 137) may include a set of patterns, each of which may have multiple states referenced to an initial quiescent state (109) for that behavioral pattern.

[0208]

[0217] Environmental Patterns One or more sensors and / or devices may be used to establish a quiescent state of the environment. This state may then be used in combination with one or more machine learning decision patterns of the behavior of one or more actors, including the PUM, to evaluate the environment being monitored. In some embodiments, this state may be monitored and recorded such that data from the sensors is not considered and / or evaluated by the devices and / or systems monitoring the environment unless an event represented by one or more datasets that changes the state of the environment is detected by the one or more sensors, devices, and / or systems. This includes a set of techniques that includes the configuration of at least one or more sensors such that data from at least one or more sensors is evaluated and / or analyzed to establish a transition from a steady state, e.g., quiescent, to a state in which the configuration of one or more sensors, devices, and / or systems is activated. The initial sensor that detects the event is called an edge sensor and may directly or indirectly initiate a change in the configuration of other sensors, devices, and / or systems that escalates the monitoring of the environment in response to the change in the state of the environment. This may include the development of one or more patterns in such a monitoring system against which incoming datasets can be compared. In some embodiments, this may include a heat map of the use of the environment by the PUM or other participant that may form part of a pattern For example, the movement of the PUM or participant as they move through the environment may form part of such a pattern, which in turn may form the basis for the configuration of one or more sensors, including edge sensors, as the person moves through the environment.

[0209]

[0218] In a steady-state situation, the focus of monitoring is broad and the data set collected from any sensor may be sparse. This collection may include random sampling of the data set for a set of devices based on variables such as time of day, the person's location in the environment, the person's behavioral patterns, etc.

[0210]

[0219] These steady states may be determined, in part, by patterns of caregiving and health-related behaviors, described herein as plateaus, where the person in caregiving has a particular health condition and therefore their particular health information, including their behavior, indicates the person's relative health status with respect to the health care plateau. For example, if a PUM has consistent conditions such as arthritis, difficulty breathing, difficulty moving, etc., these are represented by consistent patterns of behavior within the pattern and therefore described as plateaus. These health care plateaus may be defined, in part, by patterns of behavior, health datasets, health professional diagnoses, historical analysis, etc., including the use of machine learning in any configuration.

[0211]

[0220] In some embodiments, sensors, devices, and / or systems that monitor these patterns can be used to detect deviations from this plateau and variations in the data sets representative of such changes in the conditions of the PUMs and their environments. These deviations can then be used to create one or more alerts, events, and / or triggers that initiate one or more responses to these condition changes.

[0212]

[0221] In some embodiments, environmental sensing can optionally include adaptive inclusion of other devices external to the environment. For example, when a caregiver brings a medical monitoring device into the environment to perform a test on the monitored person, the generated data can be passed to the environmental monitoring system. This can result in a change in the configuration of one or more sensors, devices, and / or systems involved in the monitoring, including the external device, and can result in a change in one or more patterns operated by the system with respect to those sensors, devices, and / or systems.

[0213]

[0222] In another example, an interested party may bring a device, such as a smartphone, into the environment, and that device may become part of the monitoring device while present in the environment. In some embodiments, the device may be recognized as belonging to an interested party of the person being monitored, e.g., a friend, caregiver, etc., and thus the collected data may be transferred to the system and not permanently stored on the device in a manner that is accessible other than when the device is in the presence of the person being monitored and / or the environment. For example, the data may be tokenized using cryptographic techniques such that it can only be read under certain conditions by either the owner of the device, other devices, and / or the system.

[0214]

[0223] FIG. 8 shows an operating HCP 801) including an initial resting state (109) of that HCP and / or a set of states (Q1, P1, P2) comprising that HCP's behavioral pattern, that HCP and / or its behavioral pattern having signal processing configurations (804, 805, 806) associated with each of the patterns (802, 803) representing elements of such HCP and / or its behavioral pattern, whereby when the signal processing detects an event representing a change in the environment and / or PUM conditions, this causes a change in the operating pattern and the signal processing configuration for that pattern.

[0215]

[0224] In some embodiments, the environment may respond to the presence or absence of certain participants, for example, opening and closing doors / windows, adjusting HVAC, etc.

[0216]

[0225] In some embodiments, parties with profiles may visit the PUM, such as when friends and / or family visit the PUM at their residence or other environment. In this example, their profiles may be loaded onto the sensor, device, and / or system during their existence. Depending on the specifications of their profile and / or the PUM's profile, such data may be stored or purged from the sensor, device, and / or system, e.g., to protect their and / or the PUM's privacy.

[0217]

[0226] In some embodiments, all data storage may be encrypted using one or more key management systems. This may be true even when data is purged from the system, for example, by destruction of the key that encrypts that data and / or by further encryption and deletion of the key.

[0218]

[0227] Communication methods. Communications between sensors, devices, and systems include examples of standard communication technologies, including wireless, wired, close range or other proximity technologies, RF, LASER, visual (including optical), audio, and / or any other communication system capable of communicating between general and / or specialized devices.

[0219]

[0228] For example, Matter is an industry standard for IoT communications that attempts to unify multiple communication protocols such as Wi-Fi, Bluetooth, Zigbee, etc. onto a common IoT platform. In some embodiments, sensors, devices, and / or systems may be a set of communication methods, such as raw data communication, one or more protocols, one or more data compression schemes, etc.

[0220]

[0229] In some embodiments, the Care Village system may deploy a "Care VPN" that is used as a secure communication method between sensors, devices, and / or systems that communicate across a widely available network infrastructure.

[0221]

[0230] Since security of communications in many situations is paramount, the Care Village system may involve the use of one or more sets of tokens.

[0222]

[0231] In some embodiments, random sampling of sensor, device and / or system data feeds can be used as a monitoring technique for one or more participants and / or the environment, particularly when monitoring stationary conditions.

[0223]

[0232] In some embodiments, there may be specialized gateways, routers, data aggregators, data repositories, hubs and / or nodes that support the monitoring of one or more participants and / or environments. Such devices, described herein as care hubs, may provide one or more functions to sensors, devices, and / or systems, including one or more repositories that include elastic repository functions, processing, communication, management, configuration, aggregation, and / or other care village system management functions. Care hubs may be configured to provide different authorized and / or authenticated sensor, device, system, and / or participant identities with specific access to one or more data sets. For example, in an emergency situation, medical and / or other emergency response teams may have access to all data generated by sensors, devices, and / or systems involved in monitoring the PUM experiencing a medical emergency, including data stored in one or more elastic repositories, which may include data generated by one or more sensors, devices, and / or systems before the emergency occurs.

[0224]

[0233] In some embodiments, the care hub's communications for the environment can be configured to use different priorities and communication protocols for different sensor and / or device configurations in response to the currently operating signal processing patterns. This can include the use of secure tunnels, TLS and / or VPNs to support active medical intervention and / or prioritization of specific sensor outputs for one or more participants. The care hub can provide participants, sensors, devices and / or systems with one or more tokens that grant access to one or more data sets under one or more governing specifications.

[0225]

[0234] This care hub functionality may include segmentation of multiple sensor data sets, including tokens that represent that data to different parties with different priorities. For example, a caregiver may receive data directly in real time and have channels that include audio, video and / or other sensor data, such as motion signal processing patterns and / or the care hub has configured the sensor and communication systems to provide this information. For example, this may be when at least one event triggers a more intensive monitoring situation, such as when the person being monitored trips, drops something, etc. In this example, the caregiver may communicate directly with the monitored person and then determine if further attention is needed and in what time frame.

[0226]

[0235] In many situations, the care system uses standard off-the-shelf hardware and devices and provides configuration and control software to enable these devices to perform the required functions. Often there are standards for the operation of these devices that can be configured for this purpose. The care hub can include APIs for both ingesting data from sensors, devices, and / or systems and can provide further API access to other authorized and authenticated entities with which the care hub is communicating.

[0227]

[0236] An important aspect of the care system is respecting the privacy of those involved within it. This may be of utmost importance to PUMs, and if they feel their "personal space" is being invaded, they are less likely to accept the surveillance that could benefit them most when experiencing a transition with an HCP.

[0228]

[0237] The use of tokenization can provide an enablement for this privacy protection.

[0229]

[0238] Digital Twin The system can incorporate digital twins that are representations of the environment and the people within it. These care village digital twins (CVDTs) can be created first from an environment framework and then populated with data created by one or more sensors within that environment. This also applies to one or more parties that have a relationship with that environment in that upon instantiation as a system entity an initial profile is created and then populated with a data set. This includes the Healthcare Profiles (HCPs) of these people, specifically the PUMs, as well as the associated patterns and pattern frameworks for these people and their care.

[0230]

[0239] With enough information about multiple environments represented as digital twins, quantization and approximations can be used to represent new environments to be monitored. This provides a baseline specification of both the environment and the people within it, which can then be refined with data sets and patterns generated by sensors operating in that environment.

[0231]

[0240] This supports the use of virtual environments for research, testing, monitoring, improved detection, etc. In some embodiments, measurements and devices for the measurements can be virtualized as part of a digital twin, for example, allowing medical professionals and / or other authorized parties to access data sets generated by such devices. For example, blood pressure, oxygenation, cardiac response, etc. can be monitored by authorized parties without their presence, as long as the device is detecting PUM in the environment.

[0232]

[0241] In some embodiments, an environmental framework can be created that represents an initial course granularity of the environment to be monitored and the participants in that environment, including location, dimensions, characteristics, facilities, identities, roles and other basic data sets for the environment and participants that can include additional data sets for the people to be monitored and the participants associated with that person and the monitoring.

[0233]

[0242] Once the sensors used for monitoring have established a quiescent state of the environment, the environmental framework can be populated with data from the sensors, and the overall fidelity of the environmental framework can be increased to the point that the actual situation in the environment can be represented by the CVDTs managed and operated by the system.

[0234]

[0243] FIG. 10 illustrates the use of the care village digital twin (701) for data ingestion, state management (707), pattern identification (707, 708) and decision and / or predictive processing (703), including examples of ML / AI techniques (702).

[0235]

[0244] When both environmental and pattern data are input into the CVDT, the monitored person's care journey unfolds in relation to their HCP, which can be quantized into a set of states, whereby the relative quiescent states interspersed with the pattern are linked to provide a sequence of states over time.

[0236]

[0245] The CVDT can be used in combination with the HCP and one or more patterns for predictive purposes. For example, the CVDT may be configured with additional sensors that predict that the PUM is more likely to have a care event, such that prevention, mitigation, and / or detection of the event is more likely, and thus the environment may be augmented by the configuration of the existing one or more sensors and / or the addition of an additional one or more sensors to the environment.

[0237]

[0246] In some embodiments, at least one CVDT can identify and classify a composite feature set across one or more sensors based on the data set provided by the signal processing. These composite feature sets are sets that are initially represented in at least one CVDT and can then be passed to the configuration of sensors for monitoring the PUM and / or the environment. In some embodiments, the composite feature set includes feature arrangements that each single sensor cannot determine, but in cooperation with an appropriate care analysis management processor, a combination of sensors can identify such a composite feature set. This expands and broadens the operating limits of the sensor set to more accurately determine the behavior of the PUM in care monitoring and the environment for care monitoring. These composite feature sets can then be deployed with the operating patterns and / or sensor arrangements.

[0238]

[0247] In many situations, CVDTs are used as part of predictive analytics based on detected behavioral patterns for PUMs within the HCP. In this example, the CVDT provides a representation of the environment that is detailed enough to allow the system to evaluate the behavioral characteristics exhibited by the PUM to identify an appropriate pattern or patterns to be deployed. This may include multiple CVDTs of a single environment that are evaluated simultaneously with different patterns to identify the best match with the behavior being monitored.

[0239]

[0248] Using CVDT as a method for the representation of environments, PUMs and patterns over time can provide a visualization and representation of the care journey for a PUM, which can be used to study, analyze and / or discover behaviors, patterns or other information sets regarding a particular PUM and / or environment and / or collections of similar PUMs and / or environments to discover further patterns, care issues, etc.

[0240]

[0249] The same mechanism can be used to predict the care journey of the PUM to traverse current HCP patterns and establish the PUM's likely needs and requirements as they transition to further patterns and / or HCPs. This capability supports the creation of care regimes, financial plans, and ultimately hospice or other palliative care plans. Information generated by such CVDT analysis can be used to inform healthcare professionals supporting the PUM of specific upcoming care events, medication requirements, physical and mental limitations, etc.

[0241]

[0250] This functionality can be used for resource planning, risk management, fraud detection, financial control and impact, automatic and / or other provisioning, product and / or service facilitation and provisioning, etc.

[0242]

[0251] CVDT provides a constant representation of monitoring and can be constantly operationalized using ML / AI techniques to instantiate multiple threaded possible behaviors, patterns, and / or other situational matching.

[0243]

[0252] Such an approach would allow one or a monitoring system to move back and forth within a timeline, which can then support stakeholders, including PUMs, to have a presentation and visualization of this data, thereby enabling an assessment and understanding of a person's relative position in their HCP journey.

[0244]

[0253] In some embodiments, there may be multiple PUMs, all different but with some examples, e.g., nursing homes, similar environments. Each of these unique PUMs may have HCP behaviors that are the same or have some similar attributes, characteristics and / or features. Each of these may then have a CVDT, and the data sets from the PUMs and their environments may undergo one or more evaluations using one or more machine learning techniques. This may result, at least in part, in the identification of patterns that are applicable to all of the PUMs being monitored. In this way, new patterns may be determined and identified, including patterns that could not be identified by data from a single PUM in a single environment.

[0245]

[0254] FIG. 9 shows several PUMs (901, 911, 921), each residing in a different environment (902, 912, 922), each having HCPs (906, 916, 926) in common and / or having sufficient common attributes, and each PUM having a care village digital twin (907, 917, 927) representing that PUM, their HCPs, current behavior patterns and signal processing (905) configuration and / or state, such that one or more machine learning modules (908) can evaluate such representations to identify, determine, and classify patterns (909) that are common to all the PUMs being evaluated.

Claims

[Claim 1] A system in which at least one caregiver monitors the person receiving care, Care analysis and management processor, A plurality of environmental sensors, each of which includes at least one elastic repository configured to store a predetermined amount of dynamically configured detection data from the care recipient's environment, and connected to a computer-readable storage medium for storing the data generated by the environmental sensors, Includes, The aforementioned environmental sensor The environment of the person receiving care is detected, Determine the static state of the environment for the person receiving care, An edge state that deviates from the aforementioned stationary state and yields edge state data is detected, When the edge state is detected, the environmental sensor is configured to evaluate the data and the edge state held in the at least one elastic repository to determine whether the environmental sensor has changed to an active state. When the state changes to an active state, the environmental sensor transmits configuration specifications to at least one other sensor among the plurality of environmental sensors located in close proximity to the environmental sensor. The active environmental sensor is configured to transmit the detection data and the edge state data to the care analysis and management processor. The aforementioned care analysis and management processor A transceiver configured to receive the aforementioned detection data and the aforementioned edge status data, A microprocessor for determining whether a false positive situation has occurred, It further includes, If the aforementioned false positive situation occurs, the transceiver is configured to transmit the configuration specifications for resetting the plurality of environmental sensors to the stationary state, or If a positive result occurs, the care analysis and management processor is configured to send an alert to at least one caregiver. system.

Images