Application US20190038148

HEALTH WITH A MOBILE DEVICE
[+] Show Details

US20190038148

1. An apparatus, comprising: (8) (3)

a processing device;

a heath-indicator data sensor operatively coupled to the processing device; and

a memory having instructions stored thereon that, when executed by the processing device, cause the processing device to: receive measured low fidelity health-indicator data and other-factor data at a first time, wherein measured health-indicator data is obtained by the health-indicator data sensor;

input a set of data comprising the health-indicator data and other-factor data at the first time into a trained high-fidelity machine learning model, wherein the trained high-fidelity machine learning model generate a prediction whether a high-fidelity health-indicator signal of the user is normal or abnormal; and

in response to the prediction being abnormal, send a notification that the user's health is abnormal.

2. The apparatus according to claim 1, wherein the trained high-fidelity machine learning model comprises one or more of a trained high-fidelity generative neural network, a trained recurrent neural network (RNN), a trained feed-forward neural network, or a trained feed-forward neural network. (0)

3. The apparatus according to claim 1, wherein the trained high-fidelity machine learning model is trained on measured user health-indicator data labeled with user specific high-fidelity measurement data. (0)

4. The apparatus according to claim 1, wherein the trained high-fidelity machine learning model is trained on low fidelity health-indicator data labeled with high-fidelity measurement data, wherein the low fidelity health-indicator data and the high-fidelity measurement data is from a population of subjects. (0)

5. The apparatus according to claim 1, wherein the high-fidelity machine learning model outputs a probability distribution, wherein the prediction is sampled from the probability distribution. (2) (0)

9. The apparatus according to claim 1, wherein the mobile device is selected from the group consisting of: a smart watch; a fitness band; a computer tablet; and a laptop computer. (0)

10. The apparatus according to claim 1, wherein each of the low-fidelity health signal data and other factor-data are time segments of data over a time period. (0)

11. The apparatus according to claim 1, wherein the low-fidelity data comprises a record of heart rate prior to the first time, the other-factor data comprises a record of activity level, and the prediction of the user's health-indicator comprises a prediction that the user experienced atrial fibrillation during the record of heart rate prior to the first time. (0)

12. The apparatus according to claim 1, wherein the processing device is further to: (0) (3)

receive a set of training data, wherein the training data comprises labeled low-fidelity health-indicator data from a population of individuals, corresponding other-factor data from the population individuals, wherein the labeled low-fidelity indicator data is labeled with corresponding high-fidelity data from the population of individuals;

input an interval of the labeled low-fidelity health-indicator data and the corresponding other-factor data into the trained high-fidelity machine learning model;

update the labeled high-fidelity machine learning model by comparing an output from the trained high-fidelity machine learning model to the labeled high-fidelity data.

13. A method, comprising: (3) (3)

receiving, by a processing device, measured low fidelity health-indicator data and other-factor data at a first time, wherein measured health-indicator data is obtained by a user health-indicator data sensor;

inputting, by the processing device, data comprising the health-indicator data and other-factor data at the first time into a trained high-fidelity machine learning model, wherein the trained high-fidelity machine learning model generates a prediction whether a high-fidelity health-indicator signal of the user is normal or abnormal; and

in response to the prediction being abnormal, sending a notification that the user's health is abnormal.

14. The method according to claim 13, wherein the trained high-fidelity machine learning model comprises one or more of a trained high-fidelity generative neural network, a trained recurrent neural network (RNN), a trained feed-forward neural network, or a trained feed-forward neural network. (0)

15. The method according to claim 13, wherein each of the low-fidelity health signal data and other factor-data are time segments of data over a time period. (0)

16. The method according to claim 13, wherein the low-fidelity data comprises a record of heart rate prior to the first time, the other-factor data comprises a record of activity level, the prediction of the user's health-indicator comprises a prediction that the user experienced atrial fibrillation during the record of heart rate prior to the first time, and the notification comprises and indication to take an ECG. (0)

17. A method comprising: (2) (3)

receiving a set of training data, wherein the training data comprises a set of low-fidelity health-indicator data, a corresponding set of other-factor data, wherein the low-fidelity health-indicator data is labeled with a corresponding set of actual high-fidelity labels;

inputting, by a processing device, an interval of low-fidelity health-indicator data and corresponding other-factor data into an untrained machine learning model to generate a predicted high-fidelity label;

updating, by the processing device, the machine learning model based on comparing the predicted high-fidelity label to an actual high-fidelity label corresponding to the interval of low-fidelity health-indicator data.

18. The method of claim 17, wherein the interval of low-fidelity health-indicator data comprises heart-rate measurements over an interval of time, the other-factor data comprises activity level over the interval of time, and the actual high-fidelity label comprises an indication that atrial fibrillation occurred during that interval of time. (0)

19. The method of claim 17, further comprising: (1) (4)

inputting additional intervals of low-fidelity heath indicator data and corresponding additional other factor data into the untrained machine learning model to generate additional predicted high-fidelity labels;

comparing the additional predicted high-fidelity labels to additional actual high-fidelity labels corresponding to the additional intervals of low-fidelity health indicator data;

updating the untrained machine learning model; and

in response to determining that the untrained machine learning model has converged, outputting a trained machine learning model.

20. The method of claim 19, further comprising: (0) (3)

inputting, by the processing device, data comprising the health-indicator data and other-factor data at the first time into a trained high-fidelity machine learning model, wherein the trained high-fidelity machine learning model makes a prediction whether a high-fidelity health-indicator signal of the user is normal or abnormal; and

in response to the prediction being abnormal, sending a notification to at least the user that the user's health is abnormal.

View Abstract and Specification Size

PDF with Images and Document Face >

Full Text Publication >

Application US20190038148

HEALTH WITH A MOBILE DEVICE

Patent Matrix® Search