Glucose estimation without continuous glucose monitoring

1. A processor-implemented method comprising:

receiving input data associated with a user, the input data comprising discrete blood glucose measurement data associated with the user and contextual data associated with the user;

generating, using an estimation model and at least a portion of the input data associated with the user, one or more estimated blood glucose values associated with the user in real-time, wherein the estimation model is trained using discrete blood glucose measurement data and contextual data to determine, based at least on the contextual data associated with the user, a sequence of estimated blood glucose values associated with the user, at least a portion of the sequence of estimated blood glucose values determined without the discrete blood glucose measurement data associated with the user, and wherein the estimation model is configured to receive the discrete blood glucose measurement data associated with the user less frequently than the one or more estimated blood glucose values associated with the user are generated; and

controlling an insulin delivery device based on the generated one or more estimated blood glucose values.

2. The processor-implemented method of claim 1 , wherein the estimation model is configured to map (i) the discrete blood glucose measurement data associated with the user and the contextual data associated with the user to (ii) the sequence of estimated blood glucose values associated with the user.

3. The processor-implemented method of claim 1 , wherein:

the discrete blood glucose measurement data user used to train the estimation model is obtainable using a continuous blood glucose meter configured to measure glucose levels directly from blood; and

the discrete blood glucose measurement data associated with the user is received intermittently.

4. The processor-implemented method of claim 1 , further comprising receiving input data associated with a plurality of users within a population of users, the plurality of users comprising the user;

wherein:

the estimation model comprises a population model configured to estimate blood glucose values for the population of users; and

the population model is generated by mapping, (i) one or more of: the discrete blood glucose measurement data associated with the plurality of users or the contextual data associated with the user, to (ii) the sequence of estimated blood glucose values for the plurality of users.

5. The processor-implemented method of claim 1 , further comprising receiving input data associated with a plurality of users within a population of users, the plurality of users comprising the user;

wherein:

the estimation model comprises a customized personal model for the user, the customized personal model configured to estimate blood glucose values for the user; and

the customized personal model is generated by transforming a population model configured to estimate blood glucose values for the population of users based on mapping (i) one or more of: the discrete blood glucose measurement data associated with the user or the contextual data associated with the user, to (ii) the sequence of estimated blood glucose values for the user.

6. The processor-implemented method of claim 1 , wherein the contextual data associated with the user comprises one or more of: nutritional information about meals consumed by the user, or information about insulin delivered to the user.

7. The processor-implemented method of claim 1 , wherein the estimation model is generated by:

learning a transfer function based on one or more parameters of a machine learning model;

iteratively adjusting the one or more parameters of the machine learning model;

continuously evaluating an objective function that measures a level of mathematical agreement between estimated blood glucose values output by the machine learning model and actual measured blood glucose levels; and

generating the estimation model for the user based on the level of mathematical agreement reaching a desired threshold;

wherein the estimation model comprises the one or more parameters of the transfer function which are set to values associated with the desired threshold.

8. The processor-implemented method of claim 1 , wherein:

the discrete blood glucose measurement data associated with the user is measured using a sensor arrangement configured to provide discrete blood glucose measurements for the user; and

the contextual data includes activity data associated with the user which is provided from a source of user activity data that correlates to activity of that user.

9. The processor-implemented method of claim 1 , wherein the receiving of the input data associated with the user comprises receiving the input data via one or more input channels over a time period, the one or more input channels corresponding to at least the discrete blood glucose measurement data associated with the user obtained via a first sensor and the contextual data associated with the user.

10. A system comprising:

one or more processors; and

one or more processor-readable media storing instructions which, when executed by the one or more processors, cause performance of:

receiving input data associated with a user, the input data comprising discrete blood glucose measurement data associated with the user and contextual data associated with the user;

generating, using an estimation model and at least a portion of the input data associated with the user, one or more estimated blood glucose values associated with the user in real-time, wherein the estimation model is trained using discrete blood glucose measurement data and contextual data to determine, based at least on the contextual data associated with the user, a sequence of estimated blood glucose values associated with the user, at least a portion of the sequence of estimated blood glucose values determined without the discrete blood glucose measurement data associated with the user, and wherein the estimation model is configured to receive the discrete blood glucose measurement data associated with the user less frequently than the one or more estimated blood glucose values associated with the user are generated; and

controlling an insulin delivery device based on the generated one or more estimated blood glucose values.

11. The system of claim 10 , wherein the estimation model is configured to map (i) the discrete blood glucose measurement data associated with the user and the contextual data associated with the user to (ii) the sequence of estimated blood glucose values associated with the user.

12. The system of claim 10 , wherein:

the discrete blood glucose measurement data used to train the estimation model is obtainable using a continuous blood glucose meter configured to measure glucose levels directly from blood; and

the discrete blood glucose measurement data associated with the user is received intermittently.

13. The system of claim 10 , wherein:

the instructions, when executed by the one or more processors, cause performance of:

receiving input data associated with a plurality of users within a population of users, the plurality of users comprising the user;

the estimation model comprises a population model configured to estimate blood glucose values for the population of users; and

the population model is generated by mapping, (i) one or more of: the discrete blood glucose measurement data associated with the plurality of users or the contextual data associated with the user, to (ii) the sequence of estimated blood glucose values for the plurality of users.

14. The system of claim 10 , wherein:

the instructions, when executed by the one or more processors, cause performance of: receiving input data associated with a plurality of users within a population of users, the plurality of users comprising the user;

the estimation model comprises a customized personal model for the user, the customized personal model configured to estimate blood glucose values for the user; and

the customized personal model is generated by transforming a population model configured to estimate blood glucose values for the population of users based on mapping (i) one or more of: the discrete blood glucose measurement data associated with the user or the contextual data associated with the user, to (ii) the sequence of estimated blood glucose values for the user.

15. The system of claim 10 , wherein the estimation model is generated by:

learning a transfer function based on one or more parameters of a machine learning model;

iteratively adjusting the one or more parameters of the machine learning model;

continuously evaluating an objective function that measures a level of mathematical agreement between estimated blood glucose values output by the machine learning model and actual measured blood glucose levels; and

generating the estimation model for the user based on the level of mathematical agreement reaching a desired threshold;

wherein the estimation model comprises the one or more parameters of the transfer function which are set to values associated with the desired threshold.

16. One or more processor-readable media storing instructions which, when executed by one or more processors, cause performance of:

receiving input data associated with a user, the input data comprising discrete blood glucose measurement data associated with the user and contextual data associated with the user;

generating, using an estimation model and at least a portion of the input data associated with the user, one or more estimated blood glucose values associated with the user in real-time, wherein the estimation model is trained using discrete blood glucose measurement data and contextual data to determine, based at least on the contextual data associated with the user, a sequence of estimated blood glucose values associated with the user, at least a portion of the sequence of estimated blood glucose values determined without the discrete blood glucose measurement data associated with the user, and wherein the estimation model is configured to receive the discrete blood glucose measurement data associated with the user less frequently than the one or more estimated blood glucose values associated with the user are generated; and

controlling an insulin delivery device based on the generated one or more estimated blood glucose values.

17. The one or more processor-readable media of claim 16 , wherein the estimation model is configured to map (i) the discrete blood glucose measurement data associated with the user and the contextual data associated with the user to (ii) the sequence of estimated blood glucose values associated with the user.

18. The one or more processor-readable media of claim 16 , wherein:

the instructions, when executed by the one or more processors, cause performance of: receiving input data associated with a plurality of users within a population of users, the plurality of users comprising the user;

the estimation model comprises a population model configured to estimate blood glucose values for the population of users; and

the population model is generated by mapping, (i) one or more of: the discrete blood glucose measurement data associated with the plurality of users or the contextual data associated with the user, to (ii) the sequence of estimated blood glucose values for the plurality of users.

19. The one or more processor-readable media of claim 16 , wherein:

the instructions, when executed by the one or more processors, cause performance of: receiving input data associated with a plurality of users within a population of users, the plurality of users comprising the user;

the estimation model comprises a customized personal model for the user, the customized personal model configured to estimate blood glucose values for the user; and

the customized personal model is generated by transforming a population model configured to estimate blood glucose values for the population of users based on mapping (i) one or more of: the discrete blood glucose measurement data associated with the user or the contextual data associated with the user, to (ii) the sequence of estimated blood glucose values for the user.

20. The one or more processor-readable media of claim 16 , wherein the estimation model is generated by:

learning a transfer function based on one or more parameters of a machine learning model;

iteratively adjusting the one or more parameters of the machine learning model;

continuously evaluating an objective function that measures a level of mathematical agreement between estimated blood glucose values output by the machine learning model and actual measured blood glucose levels; and

generating the estimation model for the user based on the level of mathematical agreement reaching a desired threshold;

wherein the estimation model comprises the one or more parameters of the transfer function which are set to values associated with the desired threshold.