Machine learning-based system for estimating glucose values

Abstract

Data for a particular user can be received from a number of different input channels over a time period. The received data can include discrete blood glucose measurement data, user activity data and other contextual data for the user. The received data can be processed to generate an input data set, which can then be processed along with information from a population model, via a supervised machine learning model, to learn a transfer function for a personal model for the user that estimates blood glucose values for the user by mapping the received data to a sequence of estimated blood glucose values for the user. Parameters of the supervised machine learning model can be adjusted to generate an optimized personal model of the user that estimates blood glucose values for the user by mapping the received data to the sequence of estimated blood glucose values for the user.

What is claimed is: 
     
         1 . A method, comprising:
 receiving data for a particular user from a number of different input channels over a time period, the received data comprising: discrete blood glucose measurement data for the user; user activity data for the user; and other contextual data for the user;   processing the received data to generate an input data set;   processing the input data set and information from a population model, via a supervised machine learning model, to learn a transfer function for a personal model for the user that estimates blood glucose values for the user by mapping the received data for the over user over the time period to a sequence of estimated blood glucose values for the user over the time period, wherein the supervised machine learning model has parameters that are estimated to generate the transfer function for the population model; and   performing a parameter optimization process by adjusting one or more parameters of the supervised machine learning model to generate an optimized personal model of the user that estimates blood glucose values for the user by mapping the received data for the user over the time period to the sequence of estimated blood glucose values for the user over the time period.   
     
     
         2 . The method of  claim 1 , and wherein the supervised machine learning model is the population model, and wherein processing the input data set, comprises:
 processing the input data set, via the particular population model, to adapt the population model such that the population model is re-trained and transformed into the personal model that is personalized to the user, wherein processing the input data set comprises: learning the transfer function for the personal model of the user that estimates blood glucose values for the user by mapping the received data for the over user over the time period to a sequence of estimated blood glucose values for the user over the time period.   
     
     
         3 . The method of  claim 1 , wherein the information from the population model comprises either: other parameters that are estimated to generate a transfer function for the population model or another sequence of estimated blood glucose values for a population of users that is estimated by the transfer function for the population model, and wherein processing the input data set, via the supervised machine learning model, to learn the transfer function for the personal model, comprises:
 processing the input data set and at least some of the information from the population model, via the supervised machine learning model, to adapt a population model such that the population model is re-trained and transformed into the personal model that is personalized to the user, wherein processing the input data set comprises: learning the transfer function for the personal model of the user that estimates blood glucose values for the user by mapping the received data for the over user over the time period to a sequence of estimated blood glucose values for the user over the time period.   
     
     
         4 . The method of  claim 1 , wherein the information from the population model comprises either: other parameters that are estimated to generate a transfer function for the population model or another sequence of estimated blood glucose values for a population of users that is estimated by the transfer function for the population model, and wherein processing the input data set, via the supervised machine learning model, to learn the transfer function for the personal model, comprises:
 processing the input data set and at least some of the information from the population model, via the population model and the supervised machine learning model, to adapt the population model such that the population model is re-trained and transformed into the personal model that is personalized to the user, wherein processing the input data set comprises: learning the transfer function for the personal model of the user that estimates blood glucose values for the user by mapping the received data for the over user over the time period to a sequence of estimated blood glucose values for the user over the time period.   
     
     
         5 . The method of  claim 1 , wherein performing the parameter optimization process comprises:
 iteratively adjusting one or more parameters of the supervised machine learning model; and   continuously evaluating an objective function that measures a level of mathematical agreement between the estimated blood glucose values output by the supervised machine learning model and actual measured blood glucose levels; and   generating the optimized personal model for the user when the level of mathematical agreement between the estimated blood glucose values output by the supervised machine learning model and the actual measured blood glucose levels reaches a desired threshold, wherein the optimized personal model for the user has the parameters of the transfer function set to values that were being evaluated when the mathematical agreement reached the desired threshold.   
     
     
         6 . The method of  claim 5 , wherein the optimized personal model is an optimized window personal model, the method further comprising:
 applying a window filter to the received data, prior to processing the received data for the user, to split the received data into a series of different time windows each having a period that is less than the time period that the received data was received over, wherein each time window includes a discrete time segment of the received data; and   wherein processing the received data for the user to generate the input data set, comprises:   processing each discrete time segment of the received data to generate a corresponding input data set based on the received data received over a particular time window; and   wherein processing the input data set, comprises:   processing each corresponding input data set, via the supervised machine learning model, to learn a corresponding transfer function for a corresponding personal model that estimates blood glucose values for the user by mapping the received data for the user over the particular time window to the sequence of estimated blood glucose values for the user over the particular time window, wherein the supervised machine learning model has parameters that are estimated to generate each corresponding transfer function for each corresponding personal model.   
     
     
         7 . The method of  claim 6 , further comprising:
 selecting the corresponding personal model having the lowest error as the optimized personal model.   
     
     
         8 . The method of  claim 6 , wherein the optimized window personal model is an optimized extended window personal model, the method further comprising:
 sequentially joining each of the sequences of estimated blood glucose values for the user for each particular time window to generate a joined set of estimated blood glucose values over a number of the particular time windows to generate the optimized extended window personal model that estimates the blood glucose values for the user.   
     
     
         9 . The method of  claim 1 , wherein the discrete blood glucose measurement data for the user is measured using a sensor arrangement that provides discrete blood glucose measurements for the user,
 wherein the user activity data for the user is provided from a source of user activity data that correlates to activity of the user, and   wherein the other contextual data for the user is provided from at least one other source associated with the user and provides other information about the user.   
     
     
         10 . The method of  claim 1 , wherein each input channel comprises a different data type being measured for the user over a time period. 
     
     
         11 . A system, comprising:
 one or more hardware-based processors configured by machine-readable instructions to:   receive data for a particular user from a number of different input channels over a time period, the received data comprising: discrete blood glucose measurement data for the user; user activity data for the user; and other contextual data for the user;   process the received data to generate an input data set;   process the input data set and information from a population model, via a supervised machine learning model, to learn a transfer function for a personal model for the user that estimates blood glucose values for the user by mapping the received data for the over user over the time period to a sequence of estimated blood glucose values for the user over the time period, wherein the supervised machine learning model has parameters that are estimated to generate the transfer function for the population model; and   perform a parameter optimization process by adjusting one or more parameters of the supervised machine learning model to generate an optimized personal model of the user that estimates blood glucose values for the user by mapping the received data for the user over the time period to the sequence of estimated blood glucose values for the user over the time period.   
     
     
         12 . The system of  claim 11 , and wherein the supervised machine learning model is the population model, and wherein the one or more hardware-based processors are further configured by machine-readable instructions to:
 process the input data set, via the particular population model, to adapt the population model such that the population model is re-trained and transformed into the personal model that is personalized to the user, wherein processing the input data set comprises: learning the transfer function for the personal model of the user that estimates blood glucose values for the user by mapping the received data for the over user over the time period to a sequence of estimated blood glucose values for the user over the time period.   
     
     
         13 . The system of  claim 11 , wherein the information from the population model comprises either: other parameters that are estimated to generate a transfer function for the population model or another sequence of estimated blood glucose values for a population of users that is estimated by the transfer function for the population model, and wherein the one or more hardware-based processors are further configured by machine-readable instructions to:
 process the input data set and at least some of the information from the population model, via the supervised machine learning model, to adapt a population model such that the population model is re-trained and transformed into the personal model that is personalized to the user, wherein processing the input data set comprises: learning the transfer function for the personal model of the user that estimates blood glucose values for the user by mapping the received data for the over user over the time period to a sequence of estimated blood glucose values for the user over the time period.   
     
     
         14 . The system of  claim 11 , wherein the information from the population model comprises either: other parameters that are estimated to generate a transfer function for the population model or another sequence of estimated blood glucose values for a population of users that is estimated by the transfer function for the population model, and wherein the one or more hardware-based processors are further configured by machine-readable instructions to:
 process the input data set and at least some of the information from the population model, via the population model and the supervised machine learning model, to adapt the population model such that the population model is re-trained and transformed into the personal model that is personalized to the user, wherein processing the input data set comprises: learning the transfer function for the personal model of the user that estimates blood glucose values for the user by mapping the received data for the over user over the time period to a sequence of estimated blood glucose values for the user over the time period.   
     
     
         15 . The system of  claim 11 , wherein the one or more hardware-based processors are further configured by machine-readable instructions to:
 iteratively adjust one or more parameters of the supervised machine learning model; and   continuously evaluate an objective function that measures a level of mathematical agreement between the estimated blood glucose values output by the supervised machine learning model and actual measured blood glucose levels; and   generate the optimized personal model for the user when the level of mathematical agreement between the estimated blood glucose values output by the supervised machine learning model and the actual measured blood glucose levels reaches a desired threshold, wherein the optimized personal model for the user has the parameters of the transfer function set to values that were being evaluated when the mathematical agreement reached the desired threshold.   
     
     
         16 . The system of  claim 15 , wherein the one or more hardware-based processors are further configured by machine-readable instructions to:
 apply a window filter to the received data, prior to processing the received data for the user, to split the received data into a series of different time windows each having a period that is less than the time period that the received data was received over, wherein each time window includes a discrete time segment of the received data;   process each discrete time segment of the received data to generate a corresponding input data set based on the received data received over a particular time window; and   process each corresponding input data set, via the supervised machine learning model, to learn a corresponding transfer function for a corresponding personal model that estimates blood glucose values for the user by mapping the received data for the user over the particular time window to the sequence of estimated blood glucose values for the user over the particular time window, wherein the supervised machine learning model has parameters that are estimated to generate each corresponding transfer function for each corresponding personal model.   
     
     
         17 . The system of  claim 16 , wherein the one or more hardware-based processors are further configured by machine-readable instructions to:
 select the corresponding personal model having the lowest error as the optimized personal model.   
     
     
         18 . The system of  claim 16 , wherein the optimized window personal model is an optimized extended window personal model, and wherein the one or more hardware-based processors are further configured by machine-readable instructions to:
 sequentially join each of the sequences of estimated blood glucose values for the user for each particular time window to generate a joined set of estimated blood glucose values over a number of the particular time windows to generate the optimized extended window personal model that estimates the blood glucose values for the user.   
     
     
         19 . The system of  claim 11 , wherein the discrete blood glucose measurement data for the user is measured using a sensor arrangement that provides discrete blood glucose measurements for the user,
 wherein the user activity data for the user is provided from a source of user activity data that correlates to activity of the user, and   wherein the other contextual data for the user is provided from at least one other source associated with the user and provides other information about the user.   
     
     
         20 . At least one non-transient computer-readable medium having instructions stored thereon that are configurable to cause at least one processor to perform a method, the method comprising:
 receiving data for a particular user from a number of different input channels over a time period, the received data comprising: discrete blood glucose measurement data for the user; user activity data for the user; and other contextual data for the user;   processing the received data to generate an input data set;   processing the input data set and information from a population model, via a supervised machine learning model, to learn a transfer function for a personal model for the user that estimates blood glucose values for the user by mapping the received data for the over user over the time period to a sequence of estimated blood glucose values for the user over the time period, wherein the supervised machine learning model has parameters that are estimated to generate the transfer function for the population model; and   performing a parameter optimization process by adjusting one or more parameters of the supervised machine learning model to generate an optimized personal model of the user that estimates blood glucose values for the user by mapping the received data for the user over the time period to the sequence of estimated blood glucose values for the user over the time period.