Method and system for assessment of cognitive function based on mobile device usage
A system and method that enables a person to unobtrusively assess their cognitive function from mobile device usage. The method records on the mobile device the occurrence and timing of user events comprising the opening and closing of applications resident on the device, the characters inputted, touch-screen gestures made, and voice inputs used on those applications, performs the step of learning a function mapping from the mobile device recordings to measurements of cognitive function that uses a loss function to determine relevant features in the recording, identifies a set of optimal weights that produce a minimum of the loss function, creates a function mapping using the optimal weights, and performs the step of applying the learned function mapping to a new recording on the mobile device to compute new cognitive function values.
1. A computer-implemented means for assessing brain health of a person based on the person's use of a mobile device, wherein the computer-implemented means comprises a processing means, and wherein:
the computer-implemented means receives inputs from a monitoring means installed on the mobile device, the inputs associated with one or more interactions of the person with the mobile device, the inputs recorded by the monitoring means without requiring additional actions by the person while the inputs are recorded;
the computer-implemented means learns a function mapping from the received inputs to a brain health metric obtained from a brain test, wherein learning the function mapping uses a loss-function to determine relevant features of the inputs and to identify a set of optimal weights for the function mapping that produce a minimum of the loss-function;
the computer-implemented means receives a set of new inputs from the monitoring means, the new inputs associated with one or more new interactions of the person with the mobile device; and
the computer-implemented means applies the learned function mapping to the new inputs to calculate a new brain health metric.
2. The computer-implemented means of claim 1 , wherein the brain test is one of a neuropsychological test, a neurophysiological test, a magnetic resonance imaging (MM) test, a functional MRI test, and a positron emission tomography test.
3. The computer-implemented means of claim 1 , wherein the function mapping is one of a linear function and a non-linear function.
4. The computer-implemented means of claim 1 , wherein the one or more interactions includes at least one of: applications opened, inputs typed, gesture patterns used on a touch screen, body motions, eye movements, voice input, accelerometer sensor data, and gyroscopic sensor data.
5. The computer-implemented means of claim 1 , wherein the mobile device is one of: a smart phone, a tablet computer, and a wearable mobile device.
6. The computer-implemented means of claim 1 , wherein to quantify an effect of at least one of social interaction, mobility, physiology, cognitive stimulation, behavioral therapy, and diet on the brain health metric, the computer-implemented means:
a. learns an attribution function mapping from measurements of at least one of social interaction, mobility, physiology, cognitive stimulation, behavioral therapy, and diet to the brain health metric, the measurements associated with the inputs, wherein learning the attribution function uses a loss-function to determine relevant features of the inputs and to identify a set of optimal weights for the attribution function that produce a minimum of the loss-function and
b. outputs an attribution to the brain health metric that is explained by the measurements associated with the inputs.
7. The computer-implemented means of claim 6 , wherein the computer-implemented means makes a recommendation or care-plan adjustment in one or more of social interaction, mobility, physiology, sleep, cognitive stimulation, and diet.
8. A computer-implemented method for assessing brain health of a person based on the person's use of a mobile device, the computer-implemented method executing instructions on one or more hardware processors of a computing system, the instructions comprising the steps of:
receiving, by the computing system, inputs from a monitoring module installed on the mobile device, the inputs associated with one or more interactions of the person with the mobile device, the inputs recorded by the monitoring module without requiring additional actions by the person while the inputs are recorded;
learning, by the computing system, a function mapping from the received inputs to a brain health metric obtained from a brain test, wherein learning the function mapping uses a loss-function to determine relevant features of the inputs and to identify a set of optimal weights for the function mapping that produce a minimum of the loss-function;
receiving, by the computing system, a set of new inputs from the monitoring module, the new inputs associated with one or more new interactions of the person with the mobile device; and
applying, by the computing system, the learned function mapping to the new inputs to calculate a new brain health metric.
9. The computer-implemented method of claim 8 , wherein the brain test is one of a neuropsychological test, a neurophysiological test, a magnetic resonance imaging (MM) test, a functional MRI test, and a positron emission tomography test.
10. The computer-implemented method of claim 8 , wherein the function mapping is one of a linear function and a non-linear function.
11. The computer-implemented method of claim 8 , wherein the one or more interactions includes at least one of: applications opened, inputs typed, gesture patterns used on a touch screen, body motions, eye movements, voice input, accelerometer sensor data, and gyroscopic sensor data.
12. The computer-implemented method of claim 8 , wherein the mobile device is one of:
a smart phone, a tablet computer, and a wearable mobile device.
13. The computer-implemented method of claim 8 , wherein to quantify an effect of at least one of social interaction, mobility, physiology, cognitive stimulation, behavioral therapy, and diet on the brain health metric, the instructions further comprise the steps of:
a. learning an attribution function mapping from measurements of at least one of social interaction, mobility, physiology, cognitive stimulation, behavioral therapy, and diet to the brain health metric, the measurements associated with the inputs, wherein learning the attribution function uses a loss-function to determine relevant features of the inputs and to identify a set of optimal weights for the attribution function that produce a minimum of the loss-function; and
b. outputting an attribution to the brain health metric that is explained by the measurements associated with the inputs.
14. The computer-implemented method of claim 13 , wherein the instructions further comprise the step of making a recommendation or care-plan adjustment in one or more of social interaction, mobility, physiology, sleep, cognitive stimulation, and diet.
15. A computer-readable medium comprising instructions that when executed by a processor running on a computing system perform a method for assessing brain health of a person based on the person's use of a mobile device, the instructions comprising the steps of:
receiving, by the computing system, inputs from a monitoring module installed on the mobile device, the inputs associated with one or more interactions of the person with the mobile device, the inputs recorded by the monitoring module without requiring additional actions by the person while the inputs are recorded;
learning, by the computing system, a function mapping from the received inputs to a brain health metric obtained from a brain test, wherein learning the function mapping uses a loss-function to determine relevant features of the inputs and to identify a set of optimal weights for the learning function mapping that produce a minimum of the loss-function;
receiving, by the computing system, a set of new inputs from the monitoring module, the new inputs associated with one or more new interactions of the person with the mobile device; and
applying, by the computing system, the learned function mapping to the new inputs to calculate a new brain health metric.
16. The computer-readable medium of claim 15 , wherein the brain test is one of a neuropsychological test, a neurophysiological test, a magnetic resonance imaging (MM) test, a functional Mill test, and a positron emission tomography test.
17. The computer-readable medium of claim 15 , wherein the function mapping is one of a linear function and a non-linear function.
18. The computer-readable medium of claim 15 , wherein the one or more interactions includes at least one of: applications opened, inputs typed, gesture patterns used on a touch screen, body motions, eye movements, voice input, accelerometer sensor data, and gyroscopic sensor data.
19. The computer-readable medium of claim 15 , wherein to quantify an effect of at least one of social interaction, mobility, physiology, cognitive stimulation, behavioral therapy, and diet on the brain health metric, the instructions further comprise the steps of:
a. learning an attribution function mapping from measurements of at least one of social interaction, mobility, physiology, cognitive stimulation, behavioral therapy, and diet to the brain health metric, the measurements associated with the inputs, wherein learning the attribution function uses a loss-function to determine relevant features of the inputs and to identify a set of optimal weights for the attribution function that produce a minimum of the loss-function; and
b. outputting an attribution to the brain health metric that is explained by the measurements associated with the inputs.
20. The computer-readable medium of claim 19 , wherein the instructions further comprise the step of making a recommendation or care-plan adjustment in one or more of social interaction, mobility, physiology, sleep, cognitive stimulation, and diet.