Proxy model using mobile device data to provide health indicators
A computer-implemented method for health assessment includes providing an artificial neural network trained with training data. The method also includes receiving activity data comprising personal activity metrics corresponding to a target person, and determining, by the artificial neural network, a health indicator of the target person based upon the activity data corresponding to the target person. The method further includes outputting the health indicator of the target person to an electronic device for communication of the health indicator to a user of the electronic device. Other embodiments are disclosed.
1 . A computer-implemented method for health assessment, the method comprising:
providing an artificial neural network trained with training data, wherein the training data comprises training metrics and health indicators corresponding to training persons, wherein:
the training data is based on data from mobile electronic devices associated with the training persons; and
the artificial neural network is trained by at least identifying one or more metrics beyond a predetermined weight threshold from the training data as having one or more predictive effects on one or more of the health indicators;
receiving activity data comprising personal activity metrics corresponding to a target person, wherein:
the activity data is based on data from one or more sensors associated with the target person and associated with one or more of the training metrics, wherein (1) the one or more sensors activate to obtain sensor data associated with the one or more metrics beyond the predetermined weight threshold and (2) the one or more sensors deactivate to not obtain sensor data not associated with the one or more metrics beyond the predetermined weight threshold; and
the personal activity metrics comprise an average walking speed for the target person over a predetermined period of time;
determining, by the artificial neural network, a first health indicator and a second health indicator of the health indicators of the target person based upon the activity data corresponding to the target person, wherein:
the artificial neural network comprises a first intermediate layer to output the first health indicator, and a second intermediate layer to output the second health indicator;
the second health indicator is based on at least the first health indicator;
the first health indicator comprises at least one of a health score for the target person or a life expectancy of the target person; and
the second health indicator comprises at least one of a life insurance premium for the target person or a health insurance premium cost for the target person; and
outputting the first health indicator and the second health indicator of the target person to an electronic device for communication of the first health indicator and the second health indicator to a user of the electronic device.
2 . The computer-implemented method of claim 1 , wherein the artificial neural network is further trained by at least identifying one or more metrics not associated with the one or more metrics beyond the predetermined weight threshold from the training metrics as not having the one or more predictive effects on one or more of the health indicators.
3 . The computer-implemented method of claim 1 , wherein the health indicators are based on medical history information associated with the training persons.
4 . The computer-implemented method of claim 1 , wherein the artificial neural network is further trained with the training data by comparing a health indicator determined by the artificial neural network processing a portion of the training data corresponding to a particular training person of the training persons to a known health indicator assigned to the particular training person by a health assessment process.
5 . The computer-implemented method of claim 1 , wherein the artificial neural network comprises a recurrent neural network comprising hidden layers, and wherein the artificial neural network is trained by determining respective weights of at least one of: (i) one or more activity metrics provided as one or more inputs to the artificial neural network, or (ii) one or more outputs generated by one or more of the hidden layers.
6 . The computer-implemented method of claim 1 , wherein:
the training data is obtained from applications installed on the mobile electronic devices associated with the training persons; and
the applications are configured to obtain the sensor data associated with the one or more metrics beyond the predetermined weight threshold from sensors of the mobile electronic devices at one or more predetermined frequencies of time.
7 . The computer-implemented method of claim 1 , wherein:
the activity data associated with the target person is obtained from an application installed on at least one device associated with the target person, wherein the at least one device comprises the one or more sensors; and
the application is configured to obtain the sensor data associated with the one or more metrics beyond the predetermined weight threshold from the one or more sensors at a predetermined frequency of time.
8 . A computer system for health assessment, the computer system comprising:
one or more processors; and
one or more memories storing computer-executable instructions that, when executed by the one or more processors, cause the computing system to:
provide an artificial neural network trained with training data, wherein the training data comprises training metrics and health indicators corresponding to training persons, wherein:
the training data is based on data from mobile electronic devices associated with the training persons; and
the artificial neural network is trained by at least identifying one or more metrics beyond a predetermined weight threshold from the training data as having one or more predictive effects on one or more of the health indicators;
receive activity data comprising personal activity metrics corresponding to a target person, wherein:
the activity data is based on data from one or more sensors associated with the target person and associated with one or more of the training metrics, wherein (1) the one or more sensors activate to obtain sensor data associated with the one or more metrics beyond the predetermined weight threshold and (2) the one or more sensors deactivate to not obtain sensor data not associated with the one or more metrics beyond the predetermined weight threshold; and
the personal activity metrics comprise an average walking speed for the target person over a predetermined period of time;
determine, by the artificial neural network, a first health indicator and a second health indicator of the target person based upon the activity data corresponding to the target person, wherein:
the artificial neural network comprises a first intermediate layer to output the first health indicator, and a second intermediate layer to output the second health indicator;
the second health indicator is based on at least the first health indicator;
the first health indicator comprises at least one of a health score for the target person or a life expectancy of the target person; and
the second health indicator comprises at least one of a life insurance premium for the target person or a health insurance premium cost for the target person; and
output the first health indicator and the second health indicator of the target person to an electronic device for communication of the first health indicator and the second health indicator to a user of the electronic device.
9 . The computer system of claim 8 , wherein the artificial neural network is further trained by at least identifying one or more metrics not associated with the one or more metrics beyond the predetermined weight threshold from the training metrics as not having the one or more predictive effects on one or more of the health indicators.
10 . The computer system of claim 8 , wherein the health indicators are based on medical history information associated with the training persons.
11 . The computer system of claim 8 , wherein the artificial neural network is further trained with the training data by comparing a health indicator determined by the artificial neural network processing a portion of the training data corresponding to a particular training person of the training persons to a known health indicator assigned to the particular training person by a health assessment process.
12 . The computer system of claim 8 , wherein the artificial neural network comprises a recurrent neural network comprising hidden layers, and wherein the artificial neural network is trained by determining respective weights of at least one of: (i) one or more activity metrics provided as one or more inputs to the artificial neural network, or (ii) one or more outputs generated by one or more of the hidden layers.
13 . The computer system of claim 8 , wherein:
the training data is obtained from applications installed on the mobile electronic devices associated with the training persons; and
the applications are configured to obtain the sensor data associated with the one or more metrics beyond the predetermined weight threshold from sensors of the mobile electronic devices at one or more predetermined frequencies of time.
14 . The computer system of claim 8 , wherein:
the activity data associated with the target person is obtained from an application installed on at least one device associated with the target person, wherein the at least one device comprises the one or more sensors; and
the application is configured to obtain the sensor data associated with the one or more metrics beyond the predetermined weight threshold from the one or more sensors at a predetermined frequency of time.
15 . A non-transitory, computer-readable medium for health assessment, the non-transitory, computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to:
provide an artificial neural network trained with training data, wherein the training data comprises training metrics and health indicators corresponding to training persons, wherein:
the training data is based on data from mobile electronic devices associated with the training persons; and
the artificial neural network is trained by at least identifying one or more metrics beyond a predetermined weight threshold from the training data as having one or more predictive effects on one or more of the health indicators;
receive activity data comprising personal activity metrics corresponding to a target person, wherein:
the activity data is based on data from one or more sensors associated with the target person and associated with one or more of the training metrics, wherein (1) the one or more sensors activate to obtain sensor data associated with the one or more metrics beyond the predetermined weight threshold and (2) the one or more sensors deactivate to not obtain sensor data not associated with the one or more metrics beyond the predetermined weight threshold; and
the personal activity metrics comprise an average walking speed for the target person over a predetermined period of time;
determine, by the artificial neural network, a first health indicator and a second health indicator of the health indicators of the target person based upon the activity data corresponding to the target person, wherein:
the artificial neural network comprises a first intermediate layer to output the first health indicator, and a second intermediate layer to output the second health indicator;
the second health indicator is based on at least the first health indicator;
the first health indicator comprises at least one of a health score for the target person or a life expectancy of the target person; and
the second health indicator comprises at least one of a life insurance premium for the target person or a health insurance premium cost for the target person; and
output the first health indicator and the second health indicator of the target person to an electronic device for communication of the first health indicator and the second health indicator to a user of the electronic device.
16 . The non-transitory, computer-readable medium of claim 15 , wherein one or more of:
the artificial neural network is further trained by at least identifying one or more metrics not associated with the one or more metrics beyond the predetermined weight threshold from the training metrics as not having the one or more predictive effects on one or more of the health indicators; or
the health indicators are based on medical history information associated with the training persons.
17 . The non-transitory, computer-readable medium of claim 15 , wherein the artificial neural network is further trained with the training data by comparing a health indicator determined by the artificial neural network processing a portion of the training data corresponding to a particular training person of the training persons to a known health indicator assigned to the particular training person by a health assessment process.
18 . The non-transitory, computer-readable medium of claim 15 , wherein the artificial neural network comprises a recurrent neural network comprising hidden layers, and wherein the artificial neural network is trained by determining respective weights of at least one of: (i) one or more activity metrics provided as one or more inputs to the artificial neural network, or (ii) one or more outputs generated by one or more of the hidden layers.
19 . The non-transitory, computer-readable medium of claim 15 , wherein:
the training data is obtained from applications installed on the mobile electronic devices associated with the training persons; and
the applications are configured to obtain the sensor data associated with the one or more metrics beyond the predetermined weight threshold from sensors of the mobile electronic devices at one or more predetermined frequencies of time.
20 . The non-transitory, computer-readable medium of claim 15 , wherein:
the activity data associated with the target person is obtained from an application installed on at least one device associated with the target person, wherein the at least one device comprises the one or more sensors; and
the application is configured to obtain the sensor data associated with the one or more metrics beyond the predetermined weight threshold from the one or more sensors at a predetermined frequency of time.