Adjusting operation of a medication delivery system in response to gesture-indicated activity changes

Abstract

A system disclosed here includes an insulin infusion device, a gesture-based physical behavior detection system that generates gesture data for a user, and a controller that controls operation of the insulin infusion device. The controller performs adaptive training of at least one feature, function, setting, or model associated with the insulin infusion device, based at least in part on the sensor data. The controller processes activity-identifying data that indicates a current behavior pattern of the user, and that includes gesture data provided by the gesture-based physical behavior detection system. The controller determines, from the activity-identifying data, that the current behavior pattern differs from a currently implemented therapy behavior pattern of the user, and, in response to the determination, alters the adaptive training of the at least one feature, function, setting, or model.

What is claimed is: 
     
         1 . A method of operating a medication delivery system comprising a fluid pump mechanism, an analyte sensor to provide sensor data indicative of a physiological characteristic of a user, and at least one controller that regulates operation of the fluid pump mechanism to deliver medication from the medication delivery system to the user, based at least in part on the sensor data, the method comprising:
 performing adaptive training of at least one feature, function, setting, or model associated with the medication delivery system, based at least in part on the sensor data provided by the analyte sensor;   processing activity-identifying data that indicates a current behavior pattern of the user, the activity-identifying data comprising gesture data for the user, the gesture data provided by a gesture-based physical behavior detection system;   determining, from the activity-identifying data, that the current behavior pattern differs from a currently implemented therapy behavior pattern of the user; and   in response to the determining, altering the adaptive training of the at least one feature, function, setting, or model, resulting in an altered adaptive training scheme.   
     
     
         2 . The method of  claim 1 , further comprising:
 in response to the determining, generating a confirmation message for the user, the confirmation message requesting authorization to alter the adaptive training, wherein altering the adaptive training occurs in response to receiving an authorization to alter the adaptive training.   
     
     
         3 . The method of  claim 1 , further comprising:
 adaptively training the at least one feature, function, setting, or model in accordance with the altered adaptive training scheme for a predetermined period of time;   automatically reverting to a previous adaptive training scheme after the predetermined period of time; and   adaptively training the at least one feature, function, setting, or model in accordance with the previous adaptive training scheme.   
     
     
         4 . The method of  claim 1 , further comprising:
 adaptively training the at least one feature, function, setting, or model in accordance with the altered adaptive training scheme; and   reverting to a previous adaptive training scheme in response to receiving a user-initiated command.   
     
     
         5 . The method of  claim 1 , further comprising:
 processing updated activity-identifying data that indicates an updated behavior pattern of the user;   detecting, from the updated activity-identifying data, that the updated behavior pattern corresponds to a previously implemented therapy behavior pattern of the user; and   reverting to a previous adaptive training scheme in response to the detecting.   
     
     
         6 . The method of  claim 1 , wherein:
 the medication delivery system is controlled to automatically deliver the medication to the user in accordance with a therapy control algorithm; and   the adaptive training trains at least one therapy-altering factor of the therapy control algorithm.   
     
     
         7 . The method of  claim 6 , further comprising:
 changing one or more therapy-altering factors of the therapy control algorithm, in response to determining that the current behavior pattern differs from the currently implemented therapy behavior pattern of the user.   
     
     
         8 . The method of  claim 1 , wherein the adaptive training trains a physiological model of the user that simulates physiological response of the user to delivery of the medication. 
     
     
         9 . The method of  claim 1 , wherein altering the adaptive training occurs automatically without user input. 
     
     
         10 . The method of  claim 1 , wherein:
 the processing and determining steps are performed by a data processing system that communicates with the medication delivery system; and   the data processing system sends at least one command to the medication delivery system, the at least one command causing the medication delivery system to alter the adaptive training.   
     
     
         11 . The method of  claim 1 , wherein the processed activity-identifying data comprises user status data for the user, the user status data generated by at least one ancillary system that monitors the user. 
     
     
         12 . At least one non-transitory computer readable medium having stored thereon program code instructions that are configurable to cause at least one processor to perform a method comprising:
 performing adaptive training of at least one feature, function, setting, or model associated with a medication delivery system, based at least in part on sensor data provided by an analyte sensor that measures a physiological characteristic of a user;   processing activity-identifying data that indicates a current behavior pattern of the user, the activity-identifying data comprising gesture data for the user, the gesture data provided by a gesture-based physical behavior detection system;   determining, from the activity-identifying data, that the current behavior pattern differs from a currently implemented therapy behavior pattern of the user; and   in response to the determining, altering the adaptive training of the at least one feature, function, setting, or model, resulting in an altered adaptive training scheme.   
     
     
         13 . The at least one non-transitory computer readable medium of  claim 12 , wherein:
 the medication delivery system operates to automatically deliver the medication to the user in accordance with a therapy control algorithm; and   the adaptive training trains at least one therapy-altering factor of the therapy control algorithm.   
     
     
         14 . The at least one non-transitory computer readable medium of  claim 13 , wherein the method further comprises:
 changing one or more therapy-altering factors of the therapy control algorithm, in response to determining that the current behavior pattern differs from the currently implemented therapy behavior pattern of the user.   
     
     
         15 . The at least one non-transitory computer readable medium of  claim 12 , wherein the adaptive training trains a physiological model of the user that simulates physiological response of the user to delivery of the medication. 
     
     
         16 . The at least one non-transitory computer readable medium of  claim 12 , wherein the processed activity-identifying data comprises user status data for the user, the user status data generated by at least one ancillary system that monitors the user. 
     
     
         17 . A system comprising:
 an insulin infusion device that regulates delivery of insulin to a user;   a gesture-based physical behavior detection system configured to generate gesture data for the user, and configured to communicate the gesture data; and   at least one controller that controls operation of the insulin infusion device, the at least one controller configured to:
 perform adaptive training of at least one feature, function, setting, or model associated with the insulin infusion device, based at least in part on the sensor data; 
 process activity-identifying data that indicates a current behavior pattern of the user, the activity-identifying data comprising gesture data for the user, the gesture data provided by the gesture-based physical behavior detection system; 
 determine, from the activity-identifying data, that the current behavior pattern differs from a currently implemented therapy behavior pattern of the user; and 
 in response to the determining, altering the adaptive training of the at least one feature, function, setting, or model, resulting in an altered adaptive training scheme. 
   
     
     
         18 . The system of  claim 17 , wherein the insulin infusion device comprises the at least one controller. 
     
     
         19 . The system of  claim 17 , wherein the activity-identifying data comprises user status data for the user, the user status data generated by at least one ancillary system that monitors characteristics, status, or condition of the user. 
     
     
         20 . The system of  claim 17 , wherein:
 the insulin infusion device operates to automatically deliver insulin to the user in accordance with a therapy control algorithm; and   the adaptive training trains at least one therapy-altering factor of the therapy control algorithm.   
     
     
         21 . The system of  claim 17 , wherein the adaptive training trains a physiological model of the user that simulates physiological response of the user to delivery of insulin.