Translation modeling methods and systems for simulating sensor measurements

Abstract

Disclosed are methods and corresponding systems and devices for providing an estimation model for use with one or more instances of a particular sensor. In some aspects, an estimation model usable for estimating a value of a physiological condition is determined based at least in part on simulated measurements. The simulated measurements are generated for a first sensor, through applying a translation model to convert historical measurements associated with a second sensor into measurements that would have been produced by the first sensor. The second sensor has a different design or configuration than the first sensor. The historical measurements represent changes in the physiological condition as observed by different instances of the second sensor. The estimation model can be made available to one or more electronic devices, including at least one device configured to apply the estimation model to a measurement from a corresponding instance of the first sensor.

What is claimed is: 
     
         1 . A computer system, comprising:
 one or more processors; and   memory storing instructions which, when executed by the one or more processors, cause the computer system to perform operations comprising:
 generating simulated measurements for a first sensor based on historical measurements associated with a second sensor, wherein:
 the second sensor has a different design or configuration than the first sensor, 
 the historical measurements represent changes in a physiological condition as observed by different instances of the second sensor, and 
 generating the simulated measurements comprises applying a translation model to convert the historical measurements into measurements that would have been produced by the first sensor; 
 
 determining an estimation model using the simulated measurements, wherein the estimation model is configured to estimate a value of the physiological condition given an actual measurement from the first sensor; and 
 providing one or more electronic devices with access to the estimation model, the one or more electronic devices comprising at least one device configured to apply the estimation model to a measurement from a corresponding instance of the first sensor. 
   
     
     
         2 . The computer system of  claim 1 , wherein the operations further comprise:
 determining the translation model based at least in part on relationships between first measurements associated with the first sensor and second measurements associated with the second sensor, wherein the second measurements are separate from the historical measurements.   
     
     
         3 . The computer system of  claim 2 , wherein each of the first measurements was captured contemporaneously with a corresponding one of the second measurements. 
     
     
         4 . The computer system of  claim 1 , wherein the historical measurements correspond to values of electrical signals generated by the different instances of the second sensor in response to the physiological condition. 
     
     
         5 . The computer system of  claim 1 , wherein the estimation model is applied to convert the measurement from the corresponding instance of the first sensor into an estimated value of the physiological condition. 
     
     
         6 . The computer system of  claim 1 , wherein determining the estimation model comprises analyzing the simulated measurements together with reference values for the physiological condition. 
     
     
         7 . The computer system of  claim 6 , wherein the reference values comprise a corresponding reference measurement for each of the historical measurements. 
     
     
         8 . The computer system of  claim 6 , wherein the first sensor and the second sensor are interstitial glucose sensors, and wherein the reference values comprise blood glucose measurements. 
     
     
         9 . The computer system of  claim 6 , wherein analyzing the simulated measurements together with reference values for the physiological condition comprises determining relationships between the simulated measurements and the reference values. 
     
     
         10 . The computer system of  claim 6 , wherein determining the estimation model further comprises analyzing actual measurements from different instances of the first sensor together with corresponding reference values. 
     
     
         11 . The computer system of  claim 10 , wherein the actual measurements from different instances of the first sensor comprise measurements that were used to determine the translation model. 
     
     
         12 . The computer system of  claim 1 , wherein providing one or more electronic devices with access to the estimation model comprises sending the estimation model to the one or more electronic devices over a communications network. 
     
     
         13 . A method of providing an estimation model for use with one or more instances of a first sensor, the method comprising:
 generating, by a computer system, simulated measurements for the first sensor based on historical measurements associated with a second sensor, wherein:
 the second sensor has a different design or configuration than the first sensor, 
 the historical measurements represent changes in a physiological condition as observed by different instances of the second sensor, and 
 generating the simulated measurements comprises applying a translation model to convert the historical measurements into measurements that would have been produced by the first sensor; 
   determining the estimation model using the simulated measurements, wherein the estimation model is configured to estimate a value of the physiological condition given an actual measurement from the first sensor; and   providing, by the computer system, one or more electronic devices with access to the estimation model, the one or more electronic devices comprising at least one device configured to apply the estimation model to a measurement from a corresponding instance of the first sensor.   
     
     
         14 . The method of  claim 13 , further comprising:
 determining the translation model based at least in part on relationships between first measurements associated with the first sensor and second measurements associated with the second sensor, wherein the second measurements are separate from the historical measurements.   
     
     
         15 . The method of  claim 14 , wherein each of the first measurements was captured contemporaneously with a corresponding one of the second measurements. 
     
     
         16 . The method of  claim 13 , wherein the historical measurements correspond to values of electrical signals generated by the different instances of the second sensor in response to the physiological condition. 
     
     
         17 . The method of  claim 13 , wherein the estimation model is applied to convert the measurement from the corresponding instance of the first sensor into an estimated value of the physiological condition. 
     
     
         18 . The method of  claim 13 , wherein determining the estimation model comprises analyzing the simulated measurements together with reference values for the physiological condition, and wherein the reference values comprise a corresponding reference measurement for each of the historical measurements. 
     
     
         19 . The method of  claim 18 , wherein the first sensor and the second sensor are interstitial glucose sensors, and wherein the reference measurements are blood glucose measurements. 
     
     
         20 . The method of  claim 13 , wherein providing one or more electronic devices with access to the estimation model comprises sending the estimation model to the one or more electronic devices over a communications network.