Continuous monitoring of a user's health with a mobile device
Disclosed herein are devices, systems, methods and platforms for continuously monitoring the health status of a user, for example the cardiac health status. The present disclosure describes systems, methods, devices, software, and platforms for continuously monitoring a user's health-indicator data (for example and without limitation PPG signals, heart rate or blood pressure) from a user-device in combination with corresponding (in time) data related to factors that may impact the health-indicator (“other-factors”) to determine whether a user has normal health as judged by or compared to, for example and not by way of limitation, either (i) a group of individuals impacted by similar other-factors, or (ii) the user him/herself impacted by similar other-factors.
1 . An apparatus, comprising:
a processing device;
a low-fidelity heath-indicator data sensor operatively coupled to the processing device;
a high-fidelity health-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:
continuously receive measured low-fidelity health-indicator data at a first time, wherein the measured low-fidelity health-indicator data is obtained by the low-fidelity health-indicator data sensor;
input a set of data comprising the measured low-fidelity health-indicator data into a trained high-fidelity machine learning model, wherein the trained high-fidelity machine learning model is configured to utilize the measured low-fidelity health-indicator data to predict health-indicator data of the user at a future time, based on a low-fidelity health-indicator threshold and a first time threshold;
in response to the predicted health-indicator data of the user at the future time being outside a normal range: receive measured high-fidelity health-indicator data obtained by the high-fidelity health-indicator data sensor at the future time; and
in response to a determination that the measured high-fidelity health-indicator data obtained at the future time is inside the normal range: modify the low-fidelity health-indicator threshold in real-time to decrease a notification sensitivity.
2 . The apparatus of claim 1 , wherein the low-fidelity health-indicator threshold corresponds to a first sensitivity threshold, and wherein to modify the low-fidelity health-indicator threshold the processing device is to modify the first sensitivity threshold to a second sensitivity threshold.
3 . The apparatus of claim 2 , wherein the processing device is further to: modify the second sensitivity threshold to the first sensitivity threshold in response to an expiration of a time interval.
4 . The apparatus of claim 2 , wherein to predict the health-indicator data of the user at the future time, the processing device is to: determine whether the measured low-fidelity health-indicator data is outside the low-fidelity health-indicator threshold longer than the first time threshold.
5 . The apparatus of claim 4 , wherein the low-fidelity health-indicator threshold corresponds to the first time threshold, and wherein to modify the low-fidelity health-indicator threshold the processing device is to modify the first time threshold to a second time threshold.
6 . The apparatus of claim 1 , wherein the high-fidelity health-indicator data sensor comprises an electrocardiogram (ECG) sensor and wherein the health condition is an arrhythmia.
7 . The apparatus of claim 1 , wherein the low-fidelity health-indicator data sensor comprises a photoplethysmography (PPG) sensor.
8 . The apparatus of claim 1 , wherein the apparatus is one of: a smartwatch, a fitness band, a computer tablet, or a laptop computer.
9 . The apparatus of claim 1 , wherein the trained high-fidelity machine learning model comprises one or more of: a generative neural network, a recurrent neural network (RNN), or a feed-forward neural network.
10 . The apparatus of claim 1 , wherein the set of data further comprises a record of activity level of the user.
11 . A method, comprising:
continuously receiving measured low-fidelity health-indicator data at a first time, wherein the measured low-fidelity health-indicator data is obtained by a low-fidelity health-indicator data sensor;
inputting a set of data comprising the measured low-fidelity health-indicator data into a trained high-fidelity machine learning model, wherein the trained high-fidelity machine learning model is configured to utilize the measured low-fidelity health-indicator data to predict health-indicator data of a user at a future time, based on a low-fidelity health-indicator threshold and a first time threshold;
in response to the predicted health-indicator data of the user at the future time being outside a normal range: receiving measured high-fidelity health-indicator data obtained by a high-fidelity health-indicator data sensor at the future time; and
in response to determining that the measured high-fidelity health-indicator data obtained at the future time is inside the normal range: modifying, by a processing device, the low-fidelity health-indicator threshold in real-time to decrease a notification sensitivity.
12 . The method of claim 11 , wherein the low-fidelity health-indicator threshold corresponds to a first sensitivity threshold, and wherein to modify the low-fidelity health-indicator threshold the method further comprises: modifying the first sensitivity threshold to a second sensitivity threshold.
13 . The method of claim 12 , further comprising: modifying the second sensitivity threshold to the first sensitivity threshold in response to an expiration of a time interval.
14 . The method of claim 12 , wherein predicting the health-indicator data of the user at the future time comprises: determining whether the measured low-fidelity health-indicator data is outside the low-fidelity health-indicator threshold longer than the first time threshold.
15 . The method of claim 14 , wherein the low-fidelity health-indicator threshold corresponds to the first time threshold, and wherein modifying the low-fidelity health-indicator threshold comprises: modifying the first time threshold to a second time threshold.
16 . The method of claim 11 , wherein the high-fidelity health-indicator data sensor comprises an electrocardiogram (ECG) sensor and wherein the health condition is an arrhythmia.
17 . The method of claim 11 , wherein the low-fidelity health-indicator data sensor comprises a photoplethysmography (PPG) sensor.
18 . The method of claim 11 , wherein the processing device corresponds to one of:
a smartwatch, a fitness band, a computer tablet, or a laptop computer.
19 . The method of claim 11 , wherein the trained high-fidelity machine learning model comprises one or more of: a generative neural network, a recurrent neural network (RNN), or a feed-forward neural network.
20 . The method of claim 11 , wherein the set of data further comprises a record of activity level of the user.