Nursing village digital twinning system and method

Abstract

The physical system is represented by a digital twin. Each digital twin contains a specification of entity capabilities. Each digital twin contains physical characteristics of an entity. Each digital twin represents a state of the entity. Interaction between digital twins provides an accurate and timely representation of interaction between entities.

Description

[0001] Cross - reference to related applications

[0002] This application claims the priority of U.S. Provisional Patent Application No. 63 / 385,592, filed on November 30, 2022, the entire content of which is incorporated herein by reference. Technical Field

[0003] Aspects of the present disclosure generally relate to systems for monitoring a care recipient. Background Art

[0004] In a traditional infrastructure technology environment, a personal emergency response system (PERS) (also known as a medical emergency response system) allows people to seek help in an emergency by pressing a button.

[0005] An example system is a two-way voice communication pendant that allows people to seek help anywhere in their homes. Personal emergency response devices enable care recipients to age in place and live independently. Personal emergency response devices allow people to stay in touch with loved ones and emergency services via an existing landline. Summary of the Invention

[0006] An entity system is represented by digital twins. Each digital twin contains a specification of the entity's capabilities. Each digital twin contains the physical characteristics of the entity. Each digital twin represents the state of the entity. Interactions between digital twins provide an accurate and timely representation of interactions between entities.

[0007] One aspect includes a system for monitoring a care recipient by a stakeholder. A transceiver receives a plurality of tokens from a plurality of environmental sensors configured to monitor the care recipient. Each token at least partially includes a detected data set representing the behavior of the care recipient in the environment. Each behavior is represented by a multi-dimensional feature set that forms part of the care recipient's healthy care profile, where the tokens are encrypted using an encryption key. A non-transitory computer-readable storage medium stores a digital twin of the care recipient. The digital twin includes a dynamic tokenized representation of the static behavior of the care recipient in the environment. At least one processor analyzes the digital twin of the care recipient without analyzing the plurality of tokens received from the plurality of environmental sensors until the at least one processor receives a token indicating that a health or care event has occurred. When a health or care event occurs, the processor is configured to: decode the plurality of tokens using the encryption key, analyze the plurality of tokens, and change the state of the plurality of environmental sensors or notify the stakeholder.

[0008] At least one processor may be configured to identify potential future states of a care recipient by analyzing a digital twin.

[0009] At least one processor may be configured to identify potential future states of a care recipient by analyzing a digital twin and a plurality of tokens.

[0010] The processor may use machine learning to analyze the digital twin and the plurality of tokens.

[0011] The encryption key may be unique to the care recipient.

[0012] Changing the states of a plurality of environmental sensors may change the monitoring focus of the environmental sensors.

[0013] The monitoring focus may increase the fidelity or granularity of the environmental sensors.

[0014] The detected data set may be from a respiratory sensor or a heart rate sensor.

[0015] The transceiver may send a token to a second care system.

[0016] The second care system may be determined at least in part based on the detected data set.

[0017] The care system may be determined at least in part based on the relationship between at least one stakeholder and the care recipient.

[0018] At least one stakeholder may be determined by data regarding the relationship between at least one stakeholder and the care recipient.

[0019] On the other hand, there is provided a method for a stakeholder to monitor a care recipient. The transceiver receives a plurality of tokens from a plurality of environmental sensors configured to monitor the care recipient. Each token includes at least in part a detected data set representing the behavior of the care recipient in the environment. Each behavior is represented by a multi-dimensional feature set that forms part of the medical care profile of the care recipient, where the tokens are encrypted using an encryption key. A non-transitory computer-readable storage medium stores a digital twin of the care recipient. The digital twin includes a dynamic tokenized representation of the static behavior of the care recipient in the environment. At least one processor analyzes the digital twin of the care recipient without analyzing the plurality of tokens received from the plurality of environmental sensors until the at least one processor receives a token indicating that a health or care event has occurred. When a health or care event occurs, the processor is configured to: decode the plurality of tokens using the encryption key, analyze the plurality of tokens, and change the states of the plurality of environmental sensors or notify the stakeholder.

[0020] On the other hand, it includes a non - transitory computer - readable storage medium encoded with data and instructions. The data and instructions, when executed by a processor, cause a computing device to perform a method for monitoring a care recipient by a stakeholder. A transceiver receives a plurality of tokens from a plurality of environmental sensors configured to monitor the care recipient. Each token at least partially includes a detected data set representing the behavior of the care recipient in the environment. Each behavior is represented by a multi - dimensional feature set that forms part of the care recipient's medical care profile, where the tokens are encrypted using an encryption key. The non - transitory computer - readable storage medium stores a digital twin of the care recipient. The digital twin includes a dynamically tokenized representation of the static behavior of the care recipient in the environment. At least one processor analyzes the digital twin of the care recipient without analyzing the plurality of tokens received from the plurality of environmental sensors until the at least one processor receives a token indicating that a health or care event has occurred. When a health or care event occurs, the processor is configured to: decode the plurality of tokens using the encryption key, analyze the plurality of tokens, and change the state of the plurality of environmental sensors or notify the stakeholder. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] To better understand the nature and advantages of the present disclosure, reference should be made to the following description and the accompanying drawings. However, it should be understood that each of the provided figures is for illustrative purposes only and is not intended to define the scope of the present disclosure. Additionally, as a general rule, unless clearly contrary to the specification, elements in different figures are denoted by the same reference numerals and are generally the same or at least similar in function or purpose.

[0022] Figure 1 is a block diagram of the execution environment of the Care Village Digital Twins (CVDT).

[0023] Figure 2 is a block diagram of the CVDT for a stakeholder.

[0024] Figure 3 is a block diagram of the Care Village Digital Twins for one or more devices.

[0025] Figure 4 is a block diagram of the Care Village Digital Twins for the environment.

[0026] Figure 5 is a block diagram of the Care Village Digital Twins at different times.

[0027] Figure 6 is a block diagram of an alternative predictive Care Village Digital Twins.

[0028] Figure 7It is a block diagram of an alternative linear prediction care rural digital twin.

[0029] Figure 8 It is a block diagram of the predicted status of a person under monitoring (PUM).

[0030] Figure 9 It is a block diagram of a predictive care rural digital twin with configuration.

[0031] Figure 10 It is a block diagram of a pattern recognition, matching, and creation system.

[0032] Figure 11 It is a block diagram of a care rural digital twin system.

[0033] Figure 12 It is a block diagram of a care rural digital twin system.

[0034] Figure 13 It is a time graph of the prediction of a care rural digital twin.

[0035] Figure 14 It is a schematic diagram of the analysis of a care rural digital twin.

[0036] Figure 15 It is a schematic diagram of the simulation of a care rural digital twin.

[0037] Figure 16 It is a block diagram of the environment of a care rural digital twin.

[0038] Figure 17 It is a block diagram of a care rural digital twin CVDT mode and response system.

[0039] Figure 18 It is a block diagram of a care rural digital twin response system. Detailed implementation mode

[0040] Aspects of the present disclosure include representations of physical entities, such as devices, sensors, stakeholders, environments, and other entities that can be realized by creating digital twins. In the context of a care rural area, these digital twins are referred to as care rural digital twins (CVDTs). Each care rural digital twin can provide a representation of the physical entity it represents, including specifications, configurations, status, and any data that such an entity can generate in a specific context. This can include multiple sensors capable of providing comprehensive data sets, which can be represented by multiple care rural digital twins arranged as a composite CVDT for representing an environment including, for example, devices, sensors, stakeholders, and other entities in any arrangement.

[0041] The Nursing Village Digital Twin can represent the configuration of sensors in an environment (e.g., devices) and can take the form of a software simulation that mimics the configuration and operation of the sensors such that when incoming data (which may include and / or be based in whole or in part on facsimiles of that data) is fed to the sensors, the output of the physical sensors will be accurately produced. Such an arrangement can include a feedback mechanism such that through known and predictable configurations and sensor operations and the use of data generated by the sensors, the input to the sensors can be more accurately represented. This can also be the case when there are multiple sensors that can capture the environment and context in similar or different ways, thus more accurately capturing the input conditions for at least one of the sensors.

[0042] The Nursing Village Digital Twin can represent the state and configuration of sensors and the data sets generated by that sensor in association with the actual sensors. This ability to create, initialize, deploy, and operate such Nursing Village Digital Twins for individuals and groups of sensors provides a unique approach to the operation and management of these sensors, their configuration, the data sets they generate, and the environment and context in which they are located.

[0043] Each stakeholder, device, system, infrastructure, and other embodied element in the Nursing Village can have at least one Digital Twin representing its state.

[0044] Then, such Nursing Village Digital Twins can be updated continuously, periodically, and / or in response to events, triggers, alerts, or other real-time data, and / or can be refreshed based on any changes in the state of the embodied elements they represent.

[0045] A key aspect of this approach is to avoid imposing an unsustainable processing burden on the embodied elements operating within the Nursing Village, the management of the Nursing Village Digital Twins representing these embodiments, and the management, implementation, and / or reporting systems of the Nursing Village as a whole.

[0046] The representation of the state of each embodied element, the data they generate and / or transmit, and the state of the stakeholders is often confidential and private to those stakeholders, the devices monitoring them, and the systems and infrastructure supporting these activities.

[0047] The system's use of tokens to facilitate data communication has largely improved these issues, but the system can also adopt a series of vector representations of these tokens, thus to some extent avoiding the overly onerous and complex key management of such tokenized data.

[0048] Using a comprehensive set of Digital Twins to predict, detect, and avoid inconsistent incentives, errors, or other activities, behaviors, and outcomes provides a unique approach to solving these problems.

[0049] Nursing Village Digital Twin

[0050] In the Nursing Village, each entity in the system can have a digital twin. This digital twin is a representation of the entity. These representations can initially be frameworks that include datasets representing the entity, and over time, the entity can be populated with datasets generated by, for example, the Nursing Village system, including datasets from these entities, other sensors, behaviors, patterns, and / or other data sources.

[0051] The Nursing Village digital twin can have a state because each instance of the CVDT can represent the state of the entity represented by the Nursing Village digital twin. For example, the Nursing Village digital twin of a Person Under Monitoring (PUM) is a representation of that PUM at a specific time (a specific time or time period), where data on the PUM's state is integrated into the Nursing Village digital twin, thus providing a corresponding state for the CVDT.

[0052] When the Nursing Village digital twin receives data from the person under monitoring and their environment, such a Nursing Village digital twin can associate this data with the PUM's Healthcare Profile (HCP) and the patterns operating in these situations. This provides a current representation of the PUM's state and environment.

[0053] For example, the state of the Nursing Village digital twin at a point in time is a snapshot of the person under monitoring, their environment, and any relationships they have with stakeholders at that time. This snapshot can then be stored as a record of that state in a repository. Then, this snapshot and additional subsequent snapshots can form a history, and in some embodiments, these snapshots can be recorded according to a regular schedule and / or can be triggered automatically or manually. These snapshots can form an audit trail and be recorded in at least one distributed ledger.

[0054] One advantage of the Nursing Village digital twin is that the state, including a series of such states, can identify potential future states of the PUM and their environment through at least one prediction technique. In this way, the configuration of sensors can be determined, stakeholders can be alerted, resources can be provided in advance, and / or other activities can be carried out. The determination of such actions can be performed by the system monitoring function and / or can include manual intervention and / or assistance. For example, in some embodiments, at least one machine learning technique can be deployed to create a set of specifications, including, for example, the configuration of sensors, messages to stakeholders, and / or the PUM, etc.

[0055] For example, a care village digital twin can provide multiple predictable alternatives. For example, each of these instances, as a CVDT, can represent the expected state of the person being monitored and their environment. The system monitoring function can then compare these prediction results to determine the most likely outcome and generate specifications that can be deployed in appropriate situations.

[0056] In some embodiments, such a care village digital twin can be compared to other care village digital twins representing other PUMs with similar or identical HCPs and / or similar patterns. In this way, the prediction results (represented by the predictive care village digital twin) can be compared to the actual or historical results of the person being monitored. This comparison can be used to inform the system that is evaluating the CVDT and can thus change the prediction weight and subsequent activities of the monitoring system for that CVDT.

[0057] The care village digital twin can include multiple CVDTs of a set of physical entities in any arrangement. This can include patterns of operation and / or behavior of these entities. In the case of sensors or other configurable entities, this can include representations of these configurations, thus predicting and representing the data sets generated by such entities within the set of CVDTs.

[0058] In some embodiments, an environmental care village digital twin can include a set of sensor representations that populate the context of the physical environment and the environment within the representation. Each care village digital twin of such a sensor can include specifications for its operation, such as its performance parameters (e.g., sensitivity, capture rate, data format, other configuration parameters, etc.). In this way, the care village digital twin can assume the configuration of the sensor representation, i.e., the CVDT of the sensor, such that if an event occurs as part of the operation mode of the CVDT, the sensor will generate a data set representing that event. This configuration is a proposed or assumed configuration and may be different from the current operating configuration of the sensor. The hypothetical data set in this configuration can then be compared to the actual data set received by the physical sensor. If the data set received from the sensor is insufficient to identify the event predicted by the care village digital twin, the configuration of the physical sensor can be changed to match the assumed configuration of the CVDT of the sensor. In this way, the CVDT can operate multiple configurations of the sensor and, in some embodiments, use an optimal fit algorithm to establish a sensor configuration that matches the set of predicted events. Such an approach can be very beneficial when an event is detected by one sensor (e.g., a sharp increase in the acceleration of a device worn by a PUM) and can then be confirmed or verified by another sensor in that environment.

[0059] Care village digital twin of the person under monitoring (PUM)

[0060] In some embodiments, when monitored persons in the context of a care village are introduced to the village, they are assigned a unique identity across the system. Then, there may be certain additional datasets that form part of their initial representation. For example, these datasets can include:

[0061] Relationships - Other stakeholders they are related to (e.g., doctors, caregivers, friends and family, neighbors, pharmacies, delivery services, other services, etc.).

[0062] Health data - The initial health status that initiated the monitoring requirement, represented by their Healthcare Profile (HCP). This can include existing and current health conditions as well as a set of parameters to be monitored.

[0063] Environment - Their initial place of residence during monitoring and other locations they frequently visit. This can also include location, nearby health and care-related services, and social, cultural, and economic information associated with the location.

[0064] Care village sponsor - At least one entity (e.g., financial, insurance, VA, Medicare, etc.) that sponsors the monitored persons in the care village and the supporting characteristics (including available services, responsibilities / obligations, care rules, and other norms).

[0065] This dataset can form the initial framework of the digital twin of the monitored person. However, as the PUM undergoes monitoring, a series of datasets can populate the digital twin to make it more representative of the PUM's state. At the same time, the improved data and accuracy can be used to enrich the PUM's profile data.

[0066] In some embodiments, the care village digital twin framework can include a Healthcare Profile (HCP) representing the condition of the PUM that initiated the monitoring. This can include, for example, the specific condition(s) and / or treatment(s) the PUM is undergoing. The HCP provides a care framework determined by relevant health professionals.

[0067] Since compliance with health data protection (including HIPAA and equivalents) is often mandatory, such data can be stored in at least one repository that is subject to such mandatory protection.

[0068] Such a care village digital twin can include health data, creating a Healthcare Profile (HCP) that provides a care framework determined by relevant health professionals, i.e., a "health twin", which forms the basis for recording the health journey of the PUM and can be anonymized to form part of the care village health dataset. When interacting with the PUM, using the HCP and profile for the initial context can provide a basis for a set of touchpoints for stakeholders and the care village system.

[0069] When the Care Village Digital Twin (CVDT) is initialized, the HCPs and their models can be integrated into the CVDT by reference and / or embedding. This creates an initial state for the CVDT and can include at least one configuration (including sensing functions) of at least one device. Such a configuration can initially be used to establish the state of the person being monitored, the environment, and any stakeholders. The initial operation of the configured devices (including any sensors) is to establish the relationship between the models deployed for the HCPs and the PUM states in the environment that are relevant to those models.

[0070] One aspect is to create a "resting" state of the person being monitored and the environment, called the quiescent state, which is the expected and actual behavior of the PUM and the environment under normal conditions within the context of the appropriate HCPs.

[0071] Care Village Digital Twin of the Sensor

[0072] The Care Village sensors can have digital twins. When a sensor is initialized, the configuration and specifications of the sensor, along with its identity, location, stakeholder relationships, device and other sensor relationships, and the appropriate management and processing systems representing the environment and context of the sensor, are recorded in a repository. This forms the initial Care Village Digital Twin of the sensor, and the initial dataset of the sensor at initialization is also recorded in this digital twin. This combination provides an initial quiescent state of the sensor and the environment for the CVDT.

[0073] The Care Village Digital Twin of the sensor integrates the functions of the sensor, the configuration of the sensor, including any alternative configurations and the dataset of the sensor in the initial quiescent state, and the environment in which the sensor is deployed.

[0074] The Care Village Digital Twin representing the sensor and the relationship between the sensor and the environment provides an initial dataset from which any changes can be represented as changes to the initial state of the Care Village Digital Twin. These changes can be compared to the patterns of the expected dataset of the sensor or its collection as part of the environment. In some embodiments, there can be multiple CVDTs created based on operating the CVDT, which have data inputs representing different models and then can have different configurations applied to determine different outputs.

[0075] One aspect of the Care Village Digital Twin of the sensor or its collection is as a means of tracking sensor reliability, such as using a Byzantine algorithm. This can include the analysis of sensor behavior, such as being represented as a metric that combines the configuration and operation of the sensor with the dataset generated in the environment, for detecting faults, inaccuracies, and / or other changes in the dataset provided by the sensor.

[0076] This can include comparing the operation of the Nursing Village Digital Twin sensors with physical sensors deployed in the environment and / or deployed in a manner that establishes a reference for the sensors. This can be used to establish a measure of sensor reliability, which in turn can be used by the CVDT to evaluate the data sent by the sensors using these metrics and modify the operation of the CVDT to maintain an accurate alignment between the physical sensors and the CVDT.

[0077] Nursing Village Digital Twin of the environment

[0078] An initial framework of the environment can be established using data provided by the environment owner and / or residents, such as, for example, the PUM living in such an environment, a digital representation of any building 2D or 3D floor plan, including the location of the environment and the location of any features and fixtures (including, for example, power or other outlets, entrances and exits, furniture, facilities, and any appliances, as well as any health and wellness-related equipment) in the environment. The physical sensing of the environment by the capabilities of one or more sensors temporarily or permanently deployed in such an environment can partially or fully validate this initial framework.

[0079] One aspect is the overall environment in which the Nursing Village Digital Twin representing the environment is located. This can include the physical location and thus can also include, for example, weather and other environmental factors, both static and dynamic. For example, if a weather event, such as a heatwave, heavy rain, tornado, hurricane, etc., is predicted or experienced through the environment, the Nursing Village Digital Twin can represent these environmental changes and, in some embodiments, can be used to predict changes in the intensity of these events. This in turn can inform the response system of the likelihood of the resources and actions required for the health care of the PUM in such an environment when such a situation occurs.

[0080] The Nursing Village Digital Twin can have a variance metric, which is an incremental representation between the observed real-time entity represented by the Nursing Village Digital Twin and the entity state represented by the Nursing Village Digital Twin. For example, this can include time data, such as the time elapsed since the physical entity last updated the CVDT, the degree of variance and / or accuracy of any dataset that such an entity can provide to the CVDT, or any other attribute of such data that the entity can provide. For example, if a sensor is monitoring a single environmental aspect (e.g., temperature), the measurement time, degree of variance, and confidence or accuracy of the data can be passed as a variance metric to the Nursing Village Digital Twin of the sensor. In a more complex example, such as a sensor that captures audio, the variance metric can include a calculated value of the accuracy determined by the sensor directly and / or in combination with a signal processing system for the dataset.

[0081] As part of the initial configuration of the Nursing Village Digital Twin environment, it includes the HCP of the person being monitored. The HCP includes a set of patterns representing the behaviors and activities of the PUM, including the stationary states of these patterns. In this way, the overall stationary state of the environment and the PUM can be established within the running patterns.

[0082] The Nursing Village Digital Twin can be replicated as CVDT instances representing the actual state of the environment to represent potential changes in the patterns. These different patterns can be selected by the monitoring system as the most likely to occur based on sensor data from the CVDT representing the actual state of the environment and the PUM. Then, for example, a machine learning system can consider these changes to predict the most likely changes in the patterns. This approach can include using game theory and machine learning in combination to create a decision matrix of possible outcomes. Then, this can be used to change the configuration of sensors within any CVDT, including sensors representing the actual environment and the PUM.

[0083] In some embodiments, datasets from sensors, both physical sensors and datasets represented in one or more Nursing Village Digital Twins, can use an accumulation method such that as each dataset changes, the overall trend is accumulated over multiple time periods using multiple CVDTs, thereby predicting with different degrees of certainty the relatively likely range within which data from the physical sensors will match the data from sensors represented in one or more CVDTs.

[0084] The integration of sensors and markers that can be used by the sensors (e.g., barcodes, spray coatings, or applied location markers such as circles, numbers, geometric shapes, arrows, or other directional indicators, etc., which can use visible or invisible materials) can be included in the Nursing Village Digital Twin. In this way, the combination of sensors and markers can be used to accurately measure and calculate changes in the CVDT state. For example, if the sensor is monitoring a drawn line or other shape, such as using a material that is translucent in the visible light wavelength and from visible to infrared wavelengths, the movement of the PUM through this shape can be recorded. The same is true for pets, for example, with markers placed at a height above the pet's threshold. This method can be used for access points in different rooms within a residence to track the movement of the PUM without compromising its privacy.

[0085] In some cases, the use of the Nursing Village Digital Twin can predict the optimal location of such markers in the environment. These markers can be serialized to create unique instances. These markers can be applied to vehicles as well as outdoors.

[0086] The Care Village Digital Twin representing the actual initial state of the environment and the PUM can form a representation of the environment in a static state (i.e., "at rest"), for which other CVDTs can be used based on this initial state to represent the potential states of the environment and the PUM in different scenarios. In some embodiments, these predicted CVDTs can be used as a corpus for one or more machine learning techniques to identify the most likely future states of the PUM and the environment, which in turn can inform one or more systems (including sensor configurations in the environment, including those worn by the PUM) of the likelihood of predicted state changes.

[0087] Care Village Digital Twin of Stakeholders

[0088] Each stakeholder in the Care Village can have a Care Village Digital Twin. This includes individuals and organizations. For example, an insurance company can have a CVDT that represents the company's policies for other stakeholders (including the PUM and / or suppliers who provide goods and services to the insurance company), which are provided to the PUM, in whole or in part, directly or indirectly.

[0089] An individual as an entity can have a Care Village Digital Twin that represents their operations and activities within the scope of the Care Village. For example, the PUM, the PUM's family, friends, neighbors, etc. Additionally, each individual (e.g., a caregiver) can also have a CVDT, which is a representation of their interactions within the Care Village. In the case where a caregiver contracts with a service organization to provide skills to the PUM, that organization can also have a CVDT. In this example, the interaction with the PUM can involve the CVDTs of at least three entities, the PUM, a caregiver, and a care stakeholder organization. This can also include the CVDT of the environment, including one or more sensors and devices in the environment and any other CVDTs of entities related to the PUM.

[0090] Despite the potential complexity of the Care Village Digital Twin representing an entire organization, the CVDT of a stakeholder organization can include a set of stakeholder CVDTs, each with a counterparty, as the organization has a specific relationship with another Care Village stakeholder. For example, the PUM can have health insurance, which is instantiated as a contract, such as a smart contract, so the stakeholder organization can create a CVDT that embodies this contractual relationship and is personalized for a specific PUM. In this way, the relationship between the PUM CVDT and the stakeholder organization (e.g., an insurance company) becomes a simple one-to-one relationship. For example, the stakeholder organization CVDT can include a similar relationship with the PUM's service provider, where the stakeholder organization has a relationship with the PUM to provide these services to the PUM.

[0091] As the trend towards personalized care continues to develop, using the personal care rural digital twin relationship creates opportunities for predicting and evaluating the potential future personalized outcomes of PUM for different input sets (e.g., different care regimens, prescriptions, behavior patterns, etc.).

[0092] The use of machine learning and other big data analytics can be invoked on these data sets to identify patterns in both individuals and individual categories. These patterns can then be applied to the care rural digital twin in a way that supports predicting the possible future states of these CVDTs to determine potential outcomes and their impact on PUM healthcare.

[0093] One aspect of using the care rural digital twin in this way is the representation of the touchpoints of the various entities and their interactions. These touchpoints, which are considered interaction points, can have metrics representing these interactions such that one embodiment of these interactions can be a graph, which can then be evaluated by one or more machine learning techniques, statistical analysis, etc. to predict possible future interactions. This approach can include incentives announced and calculated by the entities involved to determine the potential range of possible outcomes.

[0094] In some embodiments, such touchpoints can be represented as a network, such as a neural network. This can also include using constraint-based logic to determine which outcomes are most likely and / or represent the set of incentives for each stakeholder, separately or overall.

[0095] In some embodiments, the care rural digital twin can be configured in a "high-touch" sampling mode, which is then monitored by the system such that when the difference between the CVDT and the real environment, as well as the entities and stakeholders within it, exceeds or approaches a threshold and / or includes an alert / event, the monitoring system is switched to the real environment. This can include changing the configuration of the entities therein to obtain a higher granularity, a higher sampling rate, a larger and more diverse data set, etc.

[0096] In some embodiments, game theory can be used to determine the strategy for data collection and / or sampling of the care rural digital twin and / or the real environment and the entities therein. For example, this can include deploying a CVDT dedicated to verifying the compliance of the monitored entities, including evaluating the incentives of the stakeholders involved.

[0097] Care Rural Digital Twin Interactions for Prediction

[0098] Each care village digital twin has specifications that can be embodied in the operational CVDT. For example, a care village digital twin can be instantiated in a computing environment based on the CVDT's specifications, and then such an instance can be operated at a time state equivalent to the actual time in the real world (i.e., network or other reference time). However, the care village digital twin can also be operated in accelerated time, where, for example, the CVDT instance operation runs at a multiple of real time or the network.

[0099] In this example, the care village digital twin can use patterns representing the health state of the monitored person in the context of the monitored person's HCP to accelerate the elapsed time. For example, using a comparison with other PUMs having the same HCP can be used to establish a future time range in which a health event is more likely to occur. This may cause the CVDT to propose a configuration variant to the PUM's sensors when anticipating such an event, thereby mitigating any negative impacts as much as possible. This can include determining and selecting the most appropriate response for such a situation.

[0100] One aspect of this approach can be to provide additional care resources, such as additional time or visits from care providers, or to alert neighbors or relatives that a health event may be more likely to occur so that they can check on or check more frequently on the monitored person. For example, a relative or other stakeholder commissioned by the PUM as an agent (e.g., having higher permissions for the configuration of at least one sensor in the PUM environment) can assume a more active monitoring role. This can include, for example, the stakeholder having an application that can identify certain aspects of the PUM's behavior (e.g., images, audio, or other sensing), and using the care village system to interpret this sensor data so that the stakeholder provides critical monitoring functions when a health event occurs or is near.

[0101] In some embodiments, the monitoring of the care village digital twin (whether operational or simulated) can be delegated to other systems, including, for example, environmental sensing, signal processing, token evaluation, incentive misalignment, or other general or specialized systems configured to monitor the real environment (including entities therein) and / or the CVDT representing such an environment. In this way, there can be specialized monitoring functions that can include human interaction optimized for specific monitoring capabilities.

[0102] The ability of the care village system to instantiate multiple care village digital twins and operate these digital twins at different time rates provides a safety ability that, for example, uses machine learning to determine the most likely outcome of at least one such CVDT. This can include those representing devices, stakeholders, systems, infrastructure, etc.

[0103] One aspect of this method is to identify, predict, extrapolate, and / or confirm at least one point of contact for each of the interacting Care Village Digital Twins, whether as an instance in a computational environment on a variable time basis or as a representation of a real-world situation.

[0104] In some embodiments, there may be a root Care Village Digital Twin that represents the current state of stakeholders, devices, environments, monitored individuals, etc. The root CVDT may also have a specification representing the state of the CVDT in a stationary state.

[0105] Then, there may be multiple instances of this Care Village Digital Twin that branch off from the root using at least one algorithm for predicting the future state of each of these branched CVDTs.

[0106] In some embodiments, different input data may be applied to multiple identical Care Village Digital Twin instances. For example, data from a previous corresponding situation may be input into at least one CVDT, and then accelerated time may be used as a prediction basis for possible situations in at least some of the CVDTs used for real-time monitoring of the PUM.

[0107] One aspect of this method may be to deploy evolutionary and genetic types of algorithms to create a set of Care Village Digital Twins with different inputs, and apply different algorithms to create a set of possible situations, which can then be used as a corpus for at least one machine learning, weighting, prediction, and / or other probabilistic techniques, including for example Bayesian techniques, to identify possible situations and outcomes of the PUM at different time scales.

[0108] Then, these variants of at least one Care Village Digital Twin can be used, for example, using different behaviors exhibited by at least one PUM with similar HCPs, in combination with an evolutionary algorithm for example, to identify and potentially monitor behaviors that may have at least one healthcare event.

[0109] This can include, for example, instantiating multiple Care Village Digital Twins, where each such Care Village Digital Twin has at least one point of contact as an interaction point with another CVDT. Then, these points of contact can be represented, for example, by a graph or other node structure, including using a manifold or other topological representation, such as a Hilbert space, etc. In these configurations, at least one machine learning technique can be applied to evaluate, determine, and / or predict the interaction for the incentives, dynamics, behaviors, and / or outcomes of the interaction.

[0110] In some embodiments, these node interactions can be represented as vectors, which can include multiple attributes and any metadata, and can thus be presented as a graphical representation interpretable by both machines and humans.

[0111] In cases where sensor data feeds are restricted and / or constrained, such as by battery or energy availability or communication capabilities, the Care Village Digital Twin can create a more continuous data set based on calculations and / or through a lookup table of previous similar data sets. This can include data revisions that work backward from, for example, a delayed data feed, such as if the person being monitored is in a vehicle between a home and a healthcare facility where communication capabilities are limited or non-existent.

[0112] Using the Care Village Digital Twin for prediction (including data sets, states, and / or interactions) can provide the ability to model, simulate, and / or emulate behavior patterns, as well as data sets that represent those patterns in any arrangement. This can include relationships between multiple Care Village Digital Twins, such as the person being monitored, caregivers, and stakeholder organizations, such that interactions between parties can be evaluated to determine which interactions have the least impact on the healthcare of the PUM or the least negative impact on all relevant parties.

[0113] Part of this evaluation is to identify and use metrics, particularly those that express profiles, contractual relationships, and / or incentives of the parties involved. In some embodiments, one aspect is the use of incentive weights in the Care Village Digital Twin to identify potential misalignments.

[0114] Each of these evaluation systems can include a configuration that supports the selection of one or more evaluation methods, such that the relationships between input data, the methods and techniques applied to that data for evaluation, and their results can be determined and / or audited. In this approach, there can also be one or more permission systems such that stakeholders can opt in or out of one or more evaluations of data sets related to their healthcare and environment.

[0115] Digital Twins and Directed Learning / Supervision

[0116] Machine learning is a subfield of artificial intelligence that focuses on computer algorithms that "learn" information directly from data using computational methods, without relying on predefined equations, sequences of instructions, and / or similar deterministic specifications. Machine learning algorithms work in a way that allows them to improve their performance as the amount of data available for learning increases. A very common way to implement machine learning is to use artificial neural networks (ANNs), which are computer systems designed to simulate the workings of the human brain by modeling the way neurons process and transmit information. A typical artificial neural network consists of multiple nodes or neurons, where each node receives multiple data inputs and produces a single data output that is a function of the weighted sum of the inputs, and these nodes are typically interconnected in a hierarchical topology, where the output of one layer of nodes serves as the input to the next layer of nodes. Artificial neural networks can have other topologies, and the way each node combines its data inputs to generate a data output can also vary.

[0117] Learning in machine learning systems based on artificial neural networks involves adjusting the weights applied to each input and the output thresholds of each node to improve the results of the overall output of the network. The learning methods in these systems are typically classified as supervised learning, unsupervised learning, and reinforcement learning. Each method is best suited for a specific set of applications.

[0118] Supervised machine learning methods achieve learning from data by providing relevant feedback. This feedback can be in the form of metadata, for example, including labels or class indicators assigned to the input data set. For example, an image of a person on the floor that is assigned the label "fall", or a combination of acceleration, height, and tilt sensor data sets that is assigned the label "walking". The feedback can also be in the form of a function that maps the input data to the desired output values. The input data and the associated metadata or output mapping are referred to as training data. The goal of supervised machine learning is to build a model that generalizes the training data to new, larger data sets.

[0119] Supervised machine learning is well-suited for classification, which refers to predicting a discrete response from input data. For example, whether a sensor data set represents a fall, a step, or other motion-related state of a PUM, whether a sound sequence represents a PUM calling for help, and / or whether a combination of risk factors of a PUM should result in calling emergency medical services, etc.

[0120] Supervised machine learning is also used in regression applications where the system predicts a continuous response from an input data set. For example, in some embodiments, a supervised machine learning system can be used to estimate a physical quantity (e.g., room temperature) to act as a virtual sensor based on historical temperature data. For example, it can be used to provide missing data for a real sensor that stops working or communicating in certain situations. This approach can also be used in other embodiments to generate simulated sensor inputs for sensors, devices, and / or environmental care rural digital twins, modeling real-life behaviors and / or situations for CVDTs.

[0121] In some embodiments, the selection and deployment of a rural digital twin for care or a set thereof can be determined at least in part to simulate or model a particular behavior, and machine learning is used to determine predictive features in a directed manner. This can include using one or more frameworks to establish results based on determined variants of the rural digital twin for care configuration. For example, for a given / desired / expected / predicted result, there may be specifications (including sets thereof) regarding the degree of allowable variation in different contexts. This can include systematic derived pattern detection of results that indicate compliance variants determined in whole or in part using machine learning techniques.

[0122] Digital twins and undirected learning / unsupervised

[0123] On the other hand, unsupervised machine learning methods do not require any labeled training data. Instead, they rely on the data itself to identify patterns and relationships. These methods can be used to identify hidden patterns and / or intrinsic structures in the input data. Unsupervised machine learning is used to cluster data points together based on common characteristics. For example, identifying pixels and / or other elements of an image belonging to an object or person in an image recognition application, or finding groups or patterns of sensor signals that are most likely to occur in a particular PUM situation. In some embodiments, clustering based on an unsupervised machine learning system can also be used to identify outliers. For example, when the pattern of a sensor data set goes beyond normal conditions, it can indicate an emergency or a sensor failure in some cases.

[0124] In some embodiments, machine learning-based clustering can also be used to identify patterns in a rural digital twin for care that lead to a particular type of result. For example, such a system can be used to identify associations between different combinations of data sets representing sensor inputs, PUM states, actions, and / or environmental states that have a desired result (avoiding falls or other emergencies, timely emergency response, etc.) or an undesired result (emergency occurs, response resources are not ready in time, notification is not provided in time, etc.).

[0125] Another classification of machine learning methods is reinforcement learning, which uses a feedback mechanism as in supervised machine learning. However, in reinforcement learning, the feedback is presented in the form of a general reward value for the generated output, rather than a set of correct output data sets. The machine learning model is typically trained through a series of trial and error until it can correctly solve each case. This approach is very useful for training a system to make decisions to achieve the desired goals in an uncertain environment. In some embodiments, the machine learning method can be combined with one or more care village digital twins, where the care village digital twins are run multiple times, and the machine learning system is trained to generate an appropriate response in the form of a decision data set to achieve the desired results of the PUM.

[0126] The combination of machine learning and game theory can provide the identification and deployment of games that represent the characteristics and behaviors of stakeholders and other entities in an environment. This is particularly useful during surveillance for detecting data inconsistencies, conflicting data sets, out-of-band data, and / or internal self-interests, etc.

[0127] One aspect of this approach is to identify real or potential contingencies, behaviors, and / or outcomes. For example, reconciling the data sets provided by one or more entities and the data sets generated by machine learning (both of which may be represented by one or more care village digital twins) can produce an assessment and reconciliation that identifies these situations, behaviors, or outcomes.

[0128] One application of directed machine learning is to identify derivatives and construct new patterns derived from system data sets, such as the patterns represented by the operation and simulation of CVDT, which match one or more characteristics of the context and PUM behavior variants.

[0129] Example embodiments

[0130] The Nursing Village Digital Twin can be instantiated as a class-like set of specifications that, when deployed in an appropriate operating environment, will become an operational CVDT with state. The Nursing Village Digital Twin includes specifications representing physical entities (e.g., devices, sensors, stakeholders, environment, etc.). These specifications include both the defined capabilities of the entity and the configuration of the entity. For example, a device can have defined specifications of the capabilities of that device, such as a thermometer, a camera with a specific focal length and resolution, etc. In the case of a stakeholder, this can include, for example, the role of the stakeholder. If the entity supports such configuration, each entity can have configuration specifications. For example, a camera can have an adjustable resolution and / or focal length, and a device including multiple sensors can have the ability to increase or decrease the sensitivity of each sensor, etc. In the case of a stakeholder, the configuration specifications can include their availability and / or working hours, cost rate, technical capabilities, etc.

[0131] Once the Nursing Village Digital Twin model is instantiated by deploying the supporting operating environment, each entity represented by the CVDT has a state that is at least partially determined by the configuration of the entity represented by the CVDT. This is the operational CVDT.

[0132] The Nursing Village Digital Twin includes the relationships between the entities represented by the CVDT. In a simple example, the CVDT can have two entities represented, a device and a stakeholder (e.g., a PUM). In this example, the device is worn by the PUM and includes three sensors, an accelerometer, a gyroscope, and an altimeter, each of which has defined and configuration specifications. The Nursing Village Digital Twin has a clear relationship between the device and the stakeholder because the device is being worn and provides a dataset regarding the state of the device and the sensors therein. This data represents the state of the device and the wearer and can form a pattern that is represented as an input set, data from the device, and an output set. This pattern can be deployed within the CVDT as additional specifications and / or can be deployed in a system that monitors the CVDT, such as a signal processing system. In either case, if the dataset changes in a way that no longer matches the specifications (including parameters, thresholds, tolerances, etc.) included in the operating pattern, including the specifications managed by the system monitoring the pattern, one or more response systems can be invoked to take action, declare an event, provide one or more configuration specifications to one or more sensors and / or the CVDT representing these entities, etc., in order to potentially provide an effective response to these dataset variations using a decision matrix.

[0133] The state of the Nursing Village Digital Twin representing these entities can be adapted based on this data, for example, by changing the configuration of these entities, etc.

[0134] This can include invoking new models representing data from entities, and / or can involve invoking additional Care Village digital twins into an operational state, and / or can involve adding additional entities to the operational CVDT in any arrangement.

[0135] Execution environment

[0136] A Care Village operating environment capable of supporting the instantiation of a Care Village digital twin specification into an operational CVDT includes a set of general operating system capabilities such as processing, preemptive scheduling, thread handling (including single-threaded and multi-threaded), storage, persistence, and repository capabilities (e.g., data management systems), as well as other standard capabilities in the form of libraries and modules. This can include a set of engines containing specific function sets, such as a physics engine, where such engines can be invoked and provided, for example, via an API and / or via a messaging system, which can be secure, for example, where the data set from the Care Village digital twin represents a set of devices, sensors, stakeholders, and / or the environment. Such engines can then apply specific operations, such as providing a model of the real-world physics, to provide an output to another system representing the physical results of the real world. The operating environment can include a set of specialized engines in any arrangement that cover various aspects of the physical world represented by the CVDT. For example, this can include a health engine, a stakeholder engine (where, for example, this can include stakeholders as organizations, where certain logic and processing can be performed on behalf of the organizational stakeholders), a CVDT prediction and simulation engine, etc.

[0137] The operating environment can be instantiated in whole or in part. For example, a subset of the functionality can be part of a device, such as a device containing a set of sensors, and can provide the lowest level of support for a Care Village digital twin representing at least one sensor. In this example, the operating environment can include communication capabilities to enable data from the Care Village digital twin and / or the sensors to be securely provided to another CVDT, which can include, for example, a device-based CVDT and a set of other CVDTs. For example, such a Care Village digital twin can contain the CVDT from the device and another set of sensors that are monitoring PUMs in the environment.

[0138] The operation support environment can be distributed, including edge computing embodiments, and can include the use of cloud-based systems, including, for example, container, serverless, and other similar cloud-based scalable deployment embodiments.

[0139] The operational support environment may include a Care Village Digital Twin Engine that provides a set of capabilities to support prediction, simulation, and other machine learning-based techniques that can be based on specifications and inputs from the operational CVDT. The engine may be called by one or more systems, such as a monitoring system, to create one or more Care Village Digital Twins that replicate the operational CVDT with different inputs, configuration specifications, and different outputs. These variants can then be considered by additional systems to establish potential trends, vectors, and / or other potential outcomes that may impact the healthcare of the PUM represented in the operational CVDT. Each of these derived CVDTs (including one or more variants of the Care Village Digital Twin from the operation) can be configured to run at an accelerated speed to determine outcomes based on changes in the input. The results can then be compared to other derived CVDT outputs with different inputs and to a direct copy of the original operational CVDT to determine a set of possible outcomes that can then be evaluated to determine the probability of such an output based on the dataset. In some cases, the configuration of sensors represented by the running Care Village Digital Twin may vary to more accurately determine if a specific predicted healthcare event has occurred or is likely to occur. For example, a camera may be configured to detect movement within a specific time range because predictions based on the dataset of the operational CVDT and the evaluation results of the predictive CVDT indicate a high likelihood of a healthcare event, such as the PUM being likely to feel dizzy when standing up from a sitting position, which could lead to a fall.

[0140] This ability to call multiple versions (including one or more direct copies) of the operational Care Village Digital Twin supports predicting and evaluating potential healthcare events without intervening in the physical PUM and its environment. This approach can be used to provide necessary resources, including predictive scheduling of those resources based on the likelihood of a healthcare event occurring to the PUM.

[0141] The dataset created during this process can be used as part of a corpus to train machine learning and AI systems, which can include the outcomes (including whether a healthcare event has occurred), to improve the overall accuracy of the prediction techniques.

[0142] Using multiple Care Village Digital Twins can include multiple branches of the CVDT with different inputs (resulting in different outputs) that can then be provided to one or more response systems (including one or more decision matrices), where the responses are evaluated by one or more evaluation systems. These response sets can be evaluated for their potential effectiveness in response to the PUM's healthcare using, for example, machine learning and / or other automated processes as well as human evaluation to determine the best response to the situation.

[0143] In this way, consideration can be invoked of possible alternatives that provide results for optimizing the health and well-being of the PUM. This provides beneficial advantages compared to fixed rule-based systems, which are typically rigid and deterministic and not sufficient to provide effective, flexible, and optimal responses.

[0144] Detect consistent variants based on operating CVDT data, such as when a person experiences a brief dizziness (detected as any head movement on an increasing basis), which subsequently leads to a fall or other health event, and then formalize this variant form of further CVDT prediction into a pattern for operating CVDT. In this way, the range of potential situations that may affect the PUM can be captured and persisted, and this way then distributes it to other PUMs with similar conditions and becomes part of a set of patterns used in the monitoring process.

[0145] In some embodiments, the operational support environment may include a stakeholder engine. This engine can provide the ability to interpret and execute specifications, such as those in the form of contracts or other structured formats, so that based on the input received by the specification engine and the processing of the specifications presented to and / or maintained by the engine, the engine will generate an output, which may include data, additional specifications, and / or configurations or other instructions for other operational support environments and / or CVDT and other entities interacting with such environments.

[0146] The specific content of the specifications will depend on the stakeholders. For example, the PUM or caregiver may have specifications, in the case of the PUM, the specifications represent their preferences, incentives, and / or constraints, while for the caregiver, they have, for example, a set of similar specifications, such as their availability, working hours, or determining factors representing their current capabilities in the caregiving area.

[0147] In an example where the stakeholder is an organization, the specifications may include contractual relationships between organizations, such as insurance companies and PUMs.

[0148] Such specifications can take the form of smart contracts, where the relationship between the stakeholder engine, the smart contract, and the distributed ledger is such that the processing of the specifications is protected (e.g., using a protected processing environment) and the security of the regulatory and control chain is ensured, so that the processing of such contracts supports the immutable records of the distributed ledger.

[0149] The stakeholder organization engine can be used to simulate potential stakeholder responses based on program specifications, such as the responses held in contracts between parties (including, for example, smart contracts), so that this data can be used in one or more response system evaluations to determine the best outcomes for the PUM and / or other counterparty stakeholders.

[0150] By providing the capabilities of a physics engine and a stakeholder engine, as well as an operating environment, the general operating environment can enable simulations, such as those that can be invoked by a CVDT engine. This can include single and multi-party simulations, where a range of predicted states and outcomes can be simulated. This can include a mixture of real and simulated results in any arrangement. Both real and simulated results can be based in whole or in part on stored datasets that have been accumulated from other stakeholders and environments that have sufficient similarity to the currently ongoing simulation. In this way, data from previous situations can be used to evaluate potential outcomes, which relate to PUMs, HCPs, and operating modes in environments similar to the current situation being simulated, with similar conditions.

[0151] In some embodiments, the simulations can provide datasets that are tested to conform to one or more sets of specifications and CVDTs representative of those specifications, to evaluate any variants. For example, it is determined that a PUM cannot be in two different locations simultaneously. Additional compliance considerations can take the form of constraints, such as when a PUM uses a specific drug, specifying the effects of that drug in the simulation to identify the effects of overdose.

[0152] One aspect of such simulations is the representation of relationships between entities, including stakeholders. For example, a fixed sensor can only be located in a single position and thus has a set of relationships with other co-located entities.

[0153] Response System

[0154] The Nursing Village Digital Twin supported by the execution environment and its engines and modules can generate datasets that are candidates for deployment in the operating CVDT, represent the physical environment including the stakeholders therein, and / or can be provided directly to sensors, devices, communication, and other systems operating in the environment represented by the CVDT.

[0155] In some embodiments, such datasets can be passed to a response system. The response system includes a set of modules that can respond to and operate on the data. The response system can include a decision support system, such as a decision matrix.

[0156] The response system results can be passed to one or more execution modules, which can operate to handle any variants in the configuration of the system and manage any dependencies, scheduling, priorities, or other process handling.

[0157] The execution modules can output specifications to the actively operating Nursing Village Digital Twin for deployment in the physical environment and / or can output specifications to one or more CVDTs to simulate the impact of the output specifications on the actual operating CVDT and the entities represented by the Nursing Village Digital Twin.

[0158] In some embodiments, the response system may include the following modules:

[0159] Communication module: The communication module provides one or more APIs capable of communicating with the CVDT to enable communication with those CVDTs and / or systems that represent and / or operate in conjunction with the CVDT. The main function of the communication system is to receive data sets from the CVDT and send data sets to the CVDT. The communication module may also be invoked by the execution module to transfer data from the execution module to the CVDT.

[0160] Execution grid: The response system may include execution capabilities in the form of an execution grid that includes a set of modules that can be configured and operated in any arrangement.

[0161] The execution grid is involved in logical processing that determines, in whole or in part, which data sets the response system can provide to the operating CVDT and / or directly to the environment represented by the CVDT. The execution grid includes a set of modules that can evaluate the received data sets and generate outputs. This may involve requesting the instantiation of additional CVDTs with different data and may include status monitoring such that the outputs of the CVDTs can be provided to any module in any arrangement.

[0162] The execution grid may include an operating environment that supports the processing of execution grid operations, including, for example, temporary and persistent storage devices, memory, processing units, machine learning systems, and other general operating support systems. In some embodiments, this may include the execution environment described herein.

[0163] Execution grid operations may use, in part or in whole, patterns, where these patterns are a set of specifications that include the configuration and / or operation of the modules managed by the execution grid operations. For example, execution grid operations may evaluate incoming data sets of one or more CVDTs by comparing them with existing data sets as part of a pattern and perform a series of configurations and instructions on the modules managed by the execution grid operation based on the comparison results.

[0164] In some embodiments, these modules include:

[0165] Decision matrix

[0166] In some embodiments, a decision matrix may be embodied as a charting system, lattice, or other formalized decision structure. For example, a multi-dimensional lattice may be used to select an appropriate data set to pass from the decision matrix to another module. The decision matrix may embody one or more outcome patterns such that traversal of the decision matrix results in consistent results based on the input data. In some cases, these deterministic results may be modified as additional data becomes available, such as when additional CVDTs of other entities (e.g., sensors) become available.

[0167] Game Theory Module (GTM)

[0168] The Game Theory Module (GTM) may employ a set of games to determine possible and appropriate outcomes of an incoming data set, thereby generating outcomes that can be passed to one or more modules.

[0169] In some embodiments, these games may be consistent with other care village games, such as those used to identify misaligned incentives. The GTM may include games determined by observing and evaluating the actions of one or more stakeholders within the care village.

[0170] The GTM may operate in conjunction with other system modules under the guidance of an execution grid to determine the appropriate games to deploy and the associated hierarchies and selections for such games. This may include games deployed in one or more CVDTs, including games instantiated by the execution grid, which may include selections deployed for one or more machine learning techniques for training and / or operating the deployment.

[0171] The execution grid may invoke one or more games for deployment for different scenarios and interactions, representing interactions between stakeholders, systems, devices, sensors, and other care village entities. Games may include cooperative and non-cooperative games, normal form and extensive form games, simultaneous and sequential move games, constant sum, zero sum, and non-zero sum games, as well as symmetric and asymmetric games, etc.

[0172] Contract Consideration Module

[0173] In some embodiments, there may be contract arrangements, such as those represented as specifications, including smart contracts that employ program logic. These specifications may be evaluated by, for example, matching and comparison systems. The evaluation of the contract under consideration may determine an appropriate data set that can be provided to one or more other modules.

[0174] In some embodiments, a contract can include a set of conditions for providing a service, product, or interaction, where the contract can represent a predicate required before an event or action occurs. The execution grid can have certain contract specifications that determine the operation of the care rural system. These specifications can be arranged in a hierarchy or priority framework such that if the conditions represented by data from a set of CVDTs represent a specific identified condition, the execution grid can take action in response to these specific conditions through one or more execution modules. For example, this can occur when the PUM is experiencing a life-threatening event, and thus the execution grid will contact 911.

[0175] Incentive Evaluation Module

[0176] In some embodiments, each stakeholder can declare and / or the system has calculated a set of incentives. These incentives can be used in whole or in part to determine the weights or other values of a data set, and / or can be represented as a data set that can be provided to one or more other modules.

[0177] In some cases, incentives can be used to change one or more CVDT configurations such that the execution grid obtains additional data sets for execution processing.

[0178] Risk Assessment Module

[0179] In some embodiments, there can be a risk assessment module that has a set of patterns represented as specifications, which represent the risks faced by stakeholders in their environment. This can include data based on, for example, the operation patterns of HCPs and / or PUMs. The data set created by this module can be provided to one or more other modules.

[0180] For example, the risk assessment module can involve considering a series of specifications such as stakeholder well-being, economy, contract commitments, service and / or product availability, etc. In some embodiments, there can be conflicting specifications from different stakeholders, and the execution grid can coordinate and / or evaluate the outputs of such specifications in one or more CVDTs instantiated for this purpose.

[0181] This can include new payment models such as usage-based insurance, etc.

[0182] In some embodiments, these modules of the execution grid can be arranged in a matrix such that each module can provide input to any other module, and each module can receive output from any other module. Then, the execution grid can use any module to select the appropriate data set to pass to one or more execution modules.

[0183] Each of these modules can be called by an execution module to provide one or more data sets to that module and / or to other modules. The execution module can then pass the resulting data set(s) to one or more execution modules using fixed or variable logic for onward transmission to one or more endpoints.

[0184] The execution module transfers data to the appropriate endpoints and manages the state of that endpoint (including any dependencies). Communication includes an acknowledgement capability such that although the communication may be asynchronous, the state of the endpoint is maintained by the execution module. This state can be reported to the execution module to maintain the overall state of the system.

[0185] One aspect of the execution module is the ability to support multiple potentially conflicting data sets simultaneously. These data sets can be compared to determine the range of data sets to be sent to the execution module under any constraints of the execution environment (e.g., a physics engine). For example, this can include configuring data sets of sensors in the environment to obtain additional data sets that can support or refute the data sets under consideration.

[0186] This evaluation can include using machine learning techniques such as deep learning, neural networks, regression modeling, etc. The execution grid can generate a set of candidate data which, in some embodiments, is then provided to the execution module. For example, this data can be provided as input to a game theory module, an incentive module, and / or a risk assessment module, each of which can then generate additional data sets for evaluation by the execution module.

[0187] Another aspect is the use of a simulation system for the care village digital twin where such simulation can provide data sets representing interventions. For example, these can be configured as CVDT-sponsored, recommended, initiated interventions which can in turn predict outcomes based on different interventions. In some embodiments, there can be configurable interventions based on these predictions which can be applied to operate the CVDT.

[0188] This intervention can be passed to a response system where, based on the operations of the execution grid, the intervention can be passed to the execution module for further transmission to the operating CVDT and / or the environment represented by the CVDT. This can include creating a set of possible interventions to determine appropriate actions by calling one or more CVDTs. For example, this can include changing the configuration of a CVDT representing an HVAC system in the environment to evaluate the outcomes for the environment and the stakeholders within it.

[0189] In some embodiments, systems such as response systems, rural care digital twin management systems, and / or other rural care systems can evaluate cost - effectiveness, cost - value, or other action - and - outcome relationships. This can include different perspectives of one or more stakeholders, such as health care, stress, ease or effort expenditure, economic or financial, etc. This can include the evaluation of state changes through the use of multiple CVDTs and their actual and / or potential outcomes based on these evaluations. This may result in potentially conflicting or other changes in the potential configurations of operating CVDTs, which can then be reviewed by the rural care system, including through the use of machine learning and / or human intervention.

[0190] Rural Care Digital Twin and Vector Tokens

[0191] One challenge for a set of sensors that potentially operate on a 24 / 7 / 365 basis is the ability of these sensors to generate continuous data sets. This is exacerbated because in many cases, there are multiple sensors, such as those in a device with multiple sensors, that are capable of generating large amounts of data, which may require very powerful processing capabilities to evaluate the data. One way to handle this data stream is to use periodic sampling to reduce the processing burden. Other strategies deployed to manage this data stream include using the evaluation of data fed at the network edge to identify action events that can trigger the data stream. Each of these methods has various limitations, including battery life, loss of critical data, dependence on the processing capabilities of the device, lack of accuracy or granularity, privacy, and security, etc. In some cases, a periodic method is often used, where data is sampled on a fixed - period basis that has been appropriately determined according to the situation.

[0192] Each of these methods has some limitations, especially the availability of feeding the raw data to another device and / or system, or the cascaded version of that data being available for other devices and / or systems.

[0193] The use of edge processing (e.g., image recognition, etc.) is typically limited by the processing capabilities of the edge device, so the degree of data evaluation can be significantly improved if the raw data is available for a processing power and support system with greater capabilities. For example, evaluation time, granularity, comparison, trigger identification, additional data integration, etc.

[0194] To address the issue of deploying appropriate processing and evaluation techniques at the required time, each edge device can be enabled and / or configured to use a vector simplification of the state of the sensor and / or device.

[0195] In some embodiments, this can include a set of vectors representing the interactions between entities (e.g., sensors and devices), where the set of vectors can be represented by one or more tokens in any arrangement.

[0196] Vectors can be represented as tokens to enable at least one system element (including signal processing, devices and / or sensors, repositories, and / or other care village system elements) to effectively distribute, manage, and use the tokens.

[0197] A vector is a simplification of data, which is represented by a token, and such a vector indicates the trajectory of a data set relative to the stationary state of a sensor.

[0198] In some embodiments, the dynamics of an operating mode can be evaluated within the boundaries of a dynamic threshold, such as a vector represented by a token generated by a sensor. For example, such boundaries can be used to determine which mode (including its set) is operating and / or will become operating or dormant. Vector analysis can use a support vector machine to establish a relevant hyperplane representing at least one decision boundary. These boundaries can be adaptive and responsive to the set of vectors being evaluated.

[0199] In some embodiments, a care village digital twin may have a set of vectors representing CVDT, particularly the trajectories of simulated or actual sensor data sets. These vectors can be passed as tokens that are protected by encryption techniques to one or more other authorized and authenticated services. Such tokens can contain identification features of sensors, devices, environments, and one or more CVDTs that contain or reference these, so that the identification can provide sufficient data about the operations of one or more systems on these entities.

[0200] In some embodiments, there may be one or more vector evaluation systems, such as support vector machines, that operate as part of an execution support environment. For example, this can include the evaluation of vectors by one or more monitoring systems, which may require access to the data in the token and / or the configuration of at least one sensor providing the data.

[0201] For example, in some embodiments, a signal processing system can receive multiple inputs from multiple sources (devices / sensors) and is configured to evaluate the combination of vectors from multiple sensors / devices, which can be represented as patterns. These patterns can then be represented by additional vectors, which are represented as tokens for further communication and evaluation.

[0202] These relationships between vectors and patterns can provide leading indicators of changes in the state of entities (including sensors, devices, environments, stakeholders, etc.), thus simplifying the evaluation process and the representation of the state of patterns and behaviors. This can include configuring multiple sensors to determine and / or confirm the accuracy of these data sets by evaluating multiple sensor data sets to improve data accuracy. For example, charts and other node arrangements can be used for both vectors and data sets in any arrangement.

[0203] In some embodiments, an entity that generates the raw data (e.g., a sensor or device) may then include the ability to represent the data as tokens that include one or more vectors.

[0204] The foregoing embodiments are provided to enable those skilled in the art to practice the present disclosure. Accordingly, the present disclosure is not intended to be limited to the embodiments shown herein, but rather should be accorded the widest scope consistent with the principles and features disclosed herein.

Claims

1. A system for a stakeholder to monitor a care recipient, comprising: A transceiver configured to receive a plurality of tokens from a plurality of environmental sensors configured to monitor the care recipient, each of the plurality of tokens at least partially comprising a detected data set representing the behavior of the care recipient in the environment, each of the behaviors being represented by a multi-dimensional feature set that forms part of the care recipient's medical care profile, wherein the tokens are encrypted using an encryption key; A non-transitory computer-readable storage medium configured to store a digital twin of the care recipient, the digital twin comprising a dynamically tokenized representation of the care recipient's static behavior in the environment; At least one processor configured to analyze the digital twin of the care recipient without analyzing the plurality of tokens received from the plurality of environmental sensors until the at least one processor receives a token indicating that a health or care event has occurred; When the health or care event occurs, the processor is configured to: Decode the plurality of tokens using the encryption key, Analyze the plurality of tokens, and Change the state of the plurality of environmental sensors or notify the stakeholder.

2. The system according to claim 1, wherein The at least one processor is further configured to identify a potential future state of the care recipient by analyzing the digital twin.

3. The system according to claim 2, wherein The at least one processor is further configured to identify a potential future state of the care recipient by analyzing the digital twin and the plurality of tokens.

4. The system according to claim 3, wherein The processor uses machine learning to analyze the digital twin and the plurality of tokens.

5. The system according to claim 4, wherein, The encryption key is unique to the care recipient.

6. The system according to claim 5, wherein Changing the state of the plurality of environmental sensors changes the monitoring focus of the environmental sensors.

7. The system according to claim 6, wherein The monitoring focus increases the fidelity or granularity of the environmental sensors.

8. The system according to claim 7, wherein, The detected data set is from a respiratory sensor or a heart rate sensor.

9. The system according to claim 1, wherein, The transceiver is configured to send the tokens to a second care system.

10. The system according to claim 9, wherein, The second care system is determined partially based on the detected data set.

11. The system according to claim 10, wherein, The second care system is determined partially based on the relationship of at least one stakeholder with the care recipient.

12. The system according to claim 11, wherein, The at least one stakeholder is determined by data regarding the relationship of the at least one stakeholder with the care recipient.

13. A method for a stakeholder to monitor a care recipient, comprising: Receiving, by a transceiver, a plurality of tokens from a plurality of environmental sensors configured to monitor the care recipient, each of the plurality of tokens at least partially comprising a detected data set representing the behavior of the care recipient in the environment, each of the behaviors being represented by a multi-dimensional feature set that forms part of the care recipient's medical care profile, wherein the tokens are encrypted using an encryption key; Store the digital twin of the cared-for person on a non-transitory computer-readable storage medium, the digital twin including a dynamically tokenized representation of the static behavior of the cared-for person in the environment; Analyze the digital twin of the cared-for person by at least one processor without analyzing the plurality of tokens received from the plurality of environmental sensors until the at least one processor receives a token indicating that a health or care event has occurred; When the health or care event occurs, the processor is configured to: Decode the plurality of tokens using the encryption key, Analyze the plurality of tokens, and Change the state of the plurality of environmental sensors or notify the stakeholders.

14. The method according to claim 13, wherein, The at least one processor is further configured to identify a potential future state of the cared-for person by analyzing the digital twin.

15. The method according to claim 14, wherein, The at least one processor is further configured to identify a potential future state of the cared-for person by analyzing the digital twin and the plurality of tokens.

16. The method according to claim 15, wherein, The processor uses machine learning to analyze the digital twin and the plurality of tokens.

17. The method according to claim 16, wherein, The encryption key is unique to the cared-for person.

18. The method according to claim 17, wherein Changing the state of the plurality of environmental sensors changes the monitoring focus of the environmental sensors.

19. A non-transitory computer-readable storage medium encoded with data and instructions that, when executed by a processor, cause a computing device to perform the following operations: Receive, via a transceiver, a plurality of tokens from a plurality of environmental sensors configured to monitor the cared-for person, each of the plurality of tokens including at least in part a detected data set representing the behavior of the cared-for person in the environment, each behavior in the behavior being represented by a multi-dimensional feature set that forms part of the medical care profile of the cared-for person, wherein the tokens are encrypted using an encryption key; Store the digital twin of the cared-for person on a non-transitory computer-readable storage medium, the digital twin including a dynamically tokenized representation of the static behavior of the cared-for person in the environment; Analyze the digital twin of the cared-for person by at least one processor without analyzing the plurality of tokens received from the plurality of environmental sensors until the at least one processor receives a token indicating that a health or care event has occurred; When the health or care event occurs, the processor is configured to: Decode the plurality of tokens using the encryption key, Analyze the plurality of tokens, and Change the state of the plurality of environmental sensors or notify the stakeholders.

20. The non-transitory computer-readable storage medium according to claim 19, wherein, The at least one processor is further configured to identify a potential future state of the cared-for person by analyzing the digital twin.

Images