Back to results
US 2019/0172579 A1Published application
Contextually Grouping Sensor Channels For Healthcare Monitoring
AssigneeVerily Life Sciences LLC
InventorsBret Edward Peterson, Dan Viet Duong
Life of this patent
Estimating the term…
Abstract
The technology described herein is directed to a health-monitoring service that facilitates contextual grouping of sensor channels for healthcare monitoring. In some implementations, clustering algorithms and predictive models can be used to dynamically select the sensor channels associated with a mobile communication device that are monitored. In this manner, only those sensor channels that provide meaningful signals are monitored.
Claims20
An apparatus comprising:
- one or more computer readable storage media storing:
- a sensor channel grouping data structure configured to cluster sensor channels into contextual groups that provide overlapping outputs,
- wherein the overlapping outputs are interchangeable for detecting that an at-risk patient or patient on a new medication or treatment has the one or more predetermined physical or emotional health-related conditions;
- program instructions that, when executed by one or more processing systems, direct the one or more processing systems to:
- identify sensor channels associated with a mobile communication device that are available for monitoring;
- select an initial subset of sensor channels from the available sensor channels based on the sensor grouping data structure;
- activate the initial subset of sensor channels as active sensor channels;
- feed outputs of the active sensor channels to a predictive model associated with the one or more health-related conditions,
- wherein the predictive model is configured to infer a health-state of a user associated with the mobile communication device based on the outputs;
- dynamically re-select, including deactivating, activating, or both, from the available sensor channels, the active sensor channels in an iterative or recursive manner based on the health-state of the user;
- and
- monitor the active sensor channels to detect if the user has the one or more predetermined physical or emotional health-related conditions.
9 more elements
An apparatus comprising:
- one or more computer readable storage media storing program instructions that, when executed by one or more processing systems, direct the one or more processing systems to:
- collect sensor data from multiple sensor channels associated with multiple communication devices;
- process the sensor data to identify training users having a predetermined health-related condition;
- correlate the sensor data associated with the training users having the predetermined health-related condition to identify sensor channels that provide information that is relevant for detecting the predetermined health-related condition;
- correlate the sensor channels that provide information that is relevant for detecting the predetermined health-related condition to identify sensor channels that provide overlapping sensor data,
- wherein the overlapping sensor data is interchangeable for detecting that an at-risk patient or patient on a new medication or treatment has the predetermined physical or emotional health-related condition;
- cluster the sensor channels that provide the overlapping sensor data into multiple sensor groups associated with the predetermined physical or emotional health-related condition;
- and
- generate a sensor grouping data structure based on the clustered sensor channels.
6 more elements
A computing system comprising:
- one or more processing systems;
- and
- one or more computer readable storage media storing:
- a sensor grouping data structure that clusters sensor channels into contextual groups providing overlapping outputs,
- wherein the overlapping outputs are interchangeable for detecting that an at-risk patient or patient on a new medication or treatment has a predetermined physical or emotional health-related condition;
- program instructions that, when executed by one or more processing systems, direct the one or more processing systems to:
- activate an initial subset of sensor channels as active sensor channels,
- wherein the initial subset of sensor channels is selected from a group of available sensor channels based on the sensor grouping data structure;
- feed outputs of the active sensor channels to a predictive model associated with the predetermined physical or emotional health-related condition,
- wherein the predictive model is configured to infer a health-state of a user associated with a mobile communication device based on the outputs;
- dynamically re-select, including deactivating, activating, or both, from the available sensor channels, the active sensor channels in an iterative or recursive manner based on the health-state of the user;
- and
- monitor the active sensor channels to detect if the user has the predetermined physical or emotional health-related condition.
10 more elements
Dependencies are inferred from the claim text.