Environment sensing for care systems

Abstract

A system, apparatus, and method to monitor 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 that environment. The person under monitoring has a care condition to be monitored, where such monitoring involves the at least two sensors providing data sets to at least one signal monitoring system. Such data sets are communicated to at least one digital twin representing the person under monitoring and their environment, such that patterns of behavior may be determined for that person. Such patterns may be represented in the at least one digital twin, as to detect behavior that indicates a change in the care condition of that person under monitoring.

Description

[00234] This care hub capability may include the segmentation of multiple sensor data sets, including tokens representing that data to differing stakeholders with differing priorities. For example, a carer may directly receive data in real time and have a channel, including audio, video and / or other sensor data where, for example an operating signal processing pattern and / or a care hub, has configured the sensors and communication systems to provide this information. For example, this may be the case where at least one event has triggered a more highly focused monitoring situation, such as when the person under monitoring has tripped, dropped something or similar. In this example, the carer may be able to have direct communication with the monitored person, and may then decide whether further attention is required and in what timeframe.

[00235] In many situations the care system will use standard off the shelf hardware and devices and provide configuration and control software to enable these devices to perform the functionality required. In many cases there are standards for operation of these devices that can be configured for this purpose. A care hub may include API’s for both ingestion of data from sensors, devices and / or systems and may provide further API access to other authorized and authenticated entities with which the care hub is in communication.

[00236] A key aspect of the care system is the respect for the privacy of the stakeholders therein. This can be of paramount importance for the PUM, where if they feel their “personal space” is being invaded they are less likely to accept the monitoring that can most benefit them as they undergo their journey in an HCP.

[00237] The use of tokenization can provide enablement for this privacy protection. Digital twins

[00238] The system may incorporate digital twins which are representations of environments and the persons therein. These Care Village Digital Twins (CVDT) can initially be created from an environment framework and then populated with data created by one or more sensors in that environment. This is also the case for one or more stakeholders that have relationships with that environment, in that an initial profile is created for them on instantiation as system entities, that is then populated with data sets. This includes the health care profiles (HCP) of those persons, specifically the PUM, and the relevant patterns and pattern frameworks for those persons and their care.

[00239] With sufficient information on multiple environments represented as digital twins, quantization and approximation may be used to represent a new environment which is to be monitored. This provides a baseline specification for both the environment and person therein, which can then be refined by the data sets and patterns generated by the sensors operating in that environment.

[00240] This supports the use of virtual environments for research, testing, monitoring and sensing improvements and the like. In some embodiments, measuring and the devices for that measuring may be virtualized as part of the digital twin, enabling, for example a health professional and / or other authorized stakeholder to access the data sets generated by such devices. For example, blood pressure, oxygenation, heart response and the like can be monitored by an authorized stakeholder without being present as long as the device is sensing the PUM at the environment.

[00241] In some embodiments, an environment framework may be created representing an initial course granularity of the environment and the stakeholders of that environment to be monitored. This may include the location, dimensions, characteristics, facilities, identity, role and other basic data sets for the environment and stakeholders which can include further data sets for the person to be monitored and the stakeholders associated with that person and the monitoring.

[00242] As the sensors employed for the monitoring establish the quiescent state for that environment, the environment framework can be populated by data from the sensors, increasing the overall fidelity of the environment framework to a point where the actual situation in the environment can be represented by a CVDT managed and operated by the system.

[00243] Figure 10 illustrates the use of care village digital twins (701) for data ingestion, state management (707), pattern identification (707,708) and determination and / or predictive processing (703), including the sue of ML / AI techniques (702).

[00244] As the CVDT is populated with both environment and pattern data, the person under monitoring whose care journey unfolds in relation to their HCP. This can be quantized into a set of states, whereby the relative quiescent states that intersperse the patterns are concatenated to provide a series of states over time.

[00245] The CVDT may be used in combination with an HCP and one or more pattern for predictive purposes. For example, the CVDT may be configured with additional sensors where a PUM is predicted to be likely to have a care incident, such that the prevention, mitigation and / or detection of that incident is more likely, and as such the environment may be augmented by the configuration of an existing one or more sensor, and / or the addition of further one or more sensors to that environment.

[00246] In some embodiments, at least one CVDT can identify and classify synthetic feature sets across one or more sensors, based on data sets provided by signal processing. These synthetic feature sets are sets that are initially represented in at least one CVDT and can then be passed to an arrangement of sensors for monitoring of a PUM and / or the environment. In some embodiments, a synthetic feature set involves features arrangements that each single sensor is not capable of determining, however in concert with an appropriate care analytics management processor a combination of sensors can identify such a synthetic feature set. This extends and expands the operating limitations of the sensor set so as to more accurately determine the behaviors of a PUM in and environment for care monitoring. These synthetic feature sets may then be deployed in operating patterns and / or sensor arrangements

[00247] In many circumstances the CVDT will be used as part of a predictive analysis based on the behavior patterns detected, for a PUM in an HCP. In this example, the CVDT provides a representation of the environment that is sufficiently detailed to enable the system to evaluate the behavioral characteristics being exhibited by a PUM, so as to ascertain the appropriate one or more patterns to be deployed. This can involve multiple CVDT of a single environment being evaluated simultaneously with differing patterns to ascertain the best match to the behaviors being monitored.

[00248] The use of CVDTs as a method for representation of the environment, PUM and patterns over a period of time can provide a visualization and representation of the care journey for the PUM. This can be used for research, analysis and / or discovery of behaviors, patterns or other information sets pertaining to the specific PUM and / or environment and / or an aggregation of similar PUM and / or environments to discover further patterns, care issues and the like.

[00249] The same mechanism can be used to forecast the care journey of a PUM to establish the likely needs and requirements of the PUM as they traverse their current HCP patterns and move to a further pattern and / or HCP. This capability supports the creation of care regimes, financial planning and eventually hospice or other palliative care planning. The information generated through such CVDT analytics may be used to inform medical practitioners supporting the PUM of specific forthcoming care incidents, medication requirements, physical and mental limitations and the like.

[00250] This capability can be used for resource planning, risk management, fraud detection, financial management and implications, automatic and / or other provisioning, product and / or service promotion and provisioning and the like.

[00251] CVDT can always be operating to provide constant representations of the monitoring and using ML / AI techniques, and can instantiate multiple threaded possible behaviors, patterns and / or other situation matching.

[00252] Such an approach allows for one or monitoring system to step back and forward in the timeline, which in turn can support stakeholders, including the PUM, having the presentation and visualization of this data so as enable evaluation and understanding of the relative position of a person in their HCP journey

[00253] In some embodiments, there may be multiple PUM, all in differing, though is some examples, such as a care facility, similar environments. Each of these unique PUM may have HCP operating that are the same or have a number of similar attributes, characteristics and / or features. Each of these may then have a CVDT and the data sets from the PUM and their environments may undergo one or more evaluations employing one or more machine learning techniques. This can result in the identification of patterns that are applicable, at least in part, to all of the PUM being monitored. In this way new patterns may be determined and identified, including those that could not have been so identified by data from a single PUM in a single environment.

[00254] Figure 9 illustrates a number of PUM (901,911,921), each domiciled in a differing environment (902,912,922), each of which HCP( 906,916,926) are common and / or have sufficient common attributes, each of which has a care village digital twin(907,917,927) representing that PUM, their HCP, current operating pattern and signal processing (905) configuration and / or state, such that one or more machine learning module (908) may evaluate such representations to identify, determine and classify patterns (909) that are common to all PUM being evaluated.

Claims

1. A system to monitor a person under care by at least one carer, comprising:a care analytics management processor;a plurality of environmental sensors, each of the environmental sensors including at least one elastic repository configured to store a dynamically configured predetermined amount of sensed data from an environment of the person under care, connected a computer-readable storage medium for storage of data generated by the environmental sensors, the environmental sensors configured to:sense the environment of the person under care,determine a quiescent state of the environment of the person under care,detect an edge condition deviating from the quiescent state resulting in edge condition data,upon detection of the edge condition, the environmental sensor is configured to evaluate the data held in the at least one elastic repository and the edge condition to determine whether the environmental sensor changes to an active state,upon changing to an active state, the environmental sensor transmits a configuration specification to at least one other sensor in the plurality of environmental sensors in proximity to the environmental sensor, andwherein the environmental sensors in the active state transmit the sensed data and the edge condition data to the care analytics management processor;the care analytics management processor, further comprising:a transceiver configured to receive the sensed data and the edge condition data, andat least one microprocessor to determine whether a false positive situation has occurred, andwhen the false positive situation has occurred, the transceiver is configured to transmit the configuration specification to reset the plurality of environmental sensors into the quiescent state, orwhen a positive situation has occurred, the care analytics management processor is configured to transmit an alert to the at least one carer.

2. The system of claim 1, wherein the transmitted configuration specification is based upon the edge condition detected by the environmental sensor.

3. The system of claim 2, wherein the transmitted configuration specification causes the at least one other sensor to dynamically invoke a threshold condition to detect the edge condition.

4. The system of claim 2, wherein the transmitted configuration specification causes the at least one other sensor to evaluate the elastic repository' to determine whether the other sensor changes to the active state.

5. The system of claim 3, wherein the transmitted configuration specification causes the at least one other sensor to evaluate the elastic repository' and the detected edge condition to determine whether the other sensor changes to the active state.

6. The system of claim 5, wherein the proximity to the environmental sensor is a physical distance.

7. The system of claim 5, wherein the proximity to the environmental sensor is alogical distance.

8. The system of claim 5, wherein the proximity to the environmental sensor is based on line-of-sight.

9. The system of claim 1, wherein the quiescent state is based on the person's physical activity level.

10. The system of claim 1, wherein the alert is a telephone call, text message, or electronic message to the carer or emergency services.

11. The system of claim 1, wherein the elastic repository records at least 30 seconds of prior data.

12. The system of claim 1, wherein the elastic repository records at least five minutes of prior data.

13. The system of claim 1, wherein the elastic repository records at least one hour of prior data.

14. The system of claim 1, wherein the care analytics management processor is a wearable sensor configured to be worn by the person.

15. The system of claim 1, wherein the care analytics management processor determines that a false negative has occurred by analyzing the sensed data and the edge condition data from more than one of the environmental sensors.

16. The system of claim 1, wherein at least one of the plurality of environmental sensors is active emission sensor device configured to create a map of the environment.

17. The system of claim 16, wherein the map of the environment is a 3dimensional model.

18. The system of claim 17, wherein the active emission sensor device is a radar device, Light Detection and Ranging (LIDAR) sensor, Radio-Frequency (RF) sensor, or Frequency-Modulated Continuous-Wave (FMCW) Radar sensor.

19. The system of claim 18, wherein the care analytics management processor is further configured to maintain an up to date and accurate positioning of objects designated by the person.

20. The system of claim 1, wherein at least one of the plurality of environmental sensors is a microphone, camera, strain gauge for impact detection, thermal sensor, motion detector, or haptic sensor.

Images