User parameter dependent cost function for personalized reduction of hypoglycemia and/or hyperglycemia in a closed loop artificial pancreas system
Exemplary embodiments described herein relate to a closed loop artificial pancreas system. The artificial pancreas system seeks to automatically and continuously control the blood glucose level of a user by emulating the endocrine functionality of a healthy pancreas. The artificial pancreas system uses a closed loop control system with a cost function. The penalty function helps to bound the infusion rate of insulin to attempt to avoid hypoglycemia and hyperglycemia. However, unlike conventional systems that use a generic or baseline parameter for a user's insulin needs in a cost function, the exemplary embodiments may use a customized parameter in the cost function that reflects the individualized insulin needs of the user. The use of the customized parameter causes the cost function to result in insulin dosages over time better suited to the individualized insulin needs of the user. This helps to better avoid hypoglycemia and hyperglycemia.
1. A device for controlling insulin deliveries to a user from an artificial pancreas, comprising:
a monitor interface with a glucose monitor to obtain glucose readings for the user from the glucose monitor;
an artificial pancreas interface for communicating with the artificial pancreas to control delivery of insulin to the user;
a processor configured to implement a control loop to control the delivery of insulin by the artificial pancreas, wherein the processor selects an insulin delivery dosage for a next delivery among delivery dosage options that has a best cost function value and wherein the cost function:
has a glucose cost component reflective of a difference between a glucose level that the dosage option is predicted to produce for the user and a target glucose level for the user,
has an insulin cost component reflective of how the dosage option differs from a current baseline insulin dosage,
has a glucose cost weight coefficient for weighting the glucose cost component,
has an insulin cost weight coefficient for weighting the insulin cost component, and
wherein at least one of the glucose cost weight coefficient and the insulin cost weight coefficient have values customized for the user.
2. The device of claim 1 , wherein the processor directs the artificial pancreas via the artificial pancreas interface to deliver the selected insulin delivery dosage.
3. The device of claim 1 , wherein only one of the glucose cost weight coefficient and the insulin cost weight coefficient has a value customized for the user.
4. The device of claim 1 , wherein both of the glucose cost weight coefficient and the insulin cost weight coefficient have values customized for the user.
5. The device of claim 1 , wherein the glucose cost weight coefficient has a value of a baseline glucose cost weight coefficient multiplied by a value indicative of a ratio of a custom value representative of insulin needs of the user to a baseline value representative of insulin needs.
6. The device of claim 5 , wherein the value indicative of the ratio is an exponential value of the ratio.
7. The device of claim 1 , wherein the insulin cost weight coefficient has a value of a baseline insulin cost weight coefficient multiplied by a value indicative of a ratio of a baseline value representative of insulin needs to a custom value representative of insulin needs of the user.
8. The device of claim 7 , wherein the value indicative of the ratio is an exponential value of the ratio.
9. The device of claim 1 , wherein the glucose cost weight coefficient has a value of a baseline glucose cost weight coefficient multiplied by a value indicative of a ratio of a baseline value representative of insulin needs to a custom value representative of insulin needs of the user.
10. The device of claim 1 , the insulin weight coefficient has a value of a baseline insulin cost weight coefficient multiplied by a value indicative ratio of a custom value representative of insulin needs of the user to a baseline value representative of insulin needs.
11. The device of claim 1 , wherein the artificial pancreas interface is a wireless communication interface.
12. The device of claim 1 , wherein the device is one of a mobile computing device, a smart phone or an insulin pump assembly.
13. The device of claim 1 , wherein the processor enforces bounds on a parameter used in determining at least one of the glucose cost weight coefficient or the insulin cost weight coefficient that is customized for the user.
14. The device of claim 13 , wherein the parameter is a value indicative of the insulin needs of the user.
15. The device of claim 1 , wherein the processor is configured to determine at least one of the glucose cost weight coefficient or the insulin cost weight coefficient based on at least one of a correction factor for insulin sensitivity for the user, an insulin to carbohydrate ratio for the user or a basal insulin level for the user.
16. The device of claim 1 , wherein at least one of the glucose cost weight coefficient and the insulin cost weight coefficient have values customized for the Total Daily Insulin (TDI) user.
17. A method performed by a processor, comprising:
receiving a glucose reading for a user from a glucose monitor;
determining a dosage for a next delivery of insulin to the user from an artificial pancreas, wherein the determining comprises:
applying a cost function to a plurality of possible dosages of insulin to the user, and
selecting a one of the possible dosages of insulin that has a best cost under the cost function;
wherein the cost function:
has a glucose cost component reflective of a difference between a glucose level that the dosage option is predicted to produce for the user and a target glucose level for the user,
has an insulin cost component reflective of how the dosage option differs from a current baseline insulin dosage,
has a glucose cost weight coefficient for weighting the glucose cost component, and
has an insulin cost weight coefficient for weighting the insulin cost component;
wherein the glucose cost weight coefficient and the insulin cost weight coefficient have values customized for the user; and
directing the artificial pancreas to deliver the selected dosage to the user.
18. A non-transitory computer-readable storage device storing computer-readable instructions that cause a processor of a device to perform the following:
receive a glucose reading for a user from a glucose monitor;
determine a dosage for a next delivery of insulin to the user from an artificial pancreas, wherein the determining comprises:
applying a cost function to a plurality of possible dosages of insulin to the user, and
selecting a one of the possible dosages of insulin that has a best cost under the cost function;
wherein the cost function:
has a glucose cost component reflective of a difference between a glucose level that the dosage option is predicted to produce for the user and a target glucose level for the user,
has an insulin cost component reflective of how the dosage option differs from a current baseline insulin dosage,
has a glucose cost weight coefficient for weighting the glucose cost component, and
has an insulin cost weight coefficient for weighting the insulin cost component;
wherein the glucose cost weight coefficient and the insulin cost weight coefficient have values customized for the user; and
direct the artificial pancreas to deliver the selected dosage to the user.
19. The non-transitory computer-readable storage medium of claim 18 , wherein only one of the glucose cost weight coefficient and the insulin cost weight coefficient has a value customized for the user.
20. The non-transitory computer-readable storage medium of claim 18 , wherein both of the glucose cost weight coefficient and the insulin cost weight coefficient have a value customized for the user.
21. The non-transitory computer-readable storage medium of claim 18 , wherein the glucose cost weight coefficient has a value of a baseline glucose cost weight coefficient multiplied by a value indicative of a ratio of a custom value representative of insulin needs of the user to a baseline value representative of insulin needs.
22. The non-transitory computer-readable storage medium of claim 21 , wherein the value indicative of the ratio is an exponential value of the ratio.
23. The non-transitory computer-readable storage medium of claim 18 , wherein the insulin cost weight coefficient has a value of a baseline insulin cost weight coefficient multiplied by a value indicative of a ratio of a baseline value representative of insulin needs to a custom value representative of insulin needs of the user.
24. The non-transitory computer-readable storage medium of claim 23 , wherein the value indicative of the ratio is an exponential value of the ratio.
25. The non-transitory computer-readable storage medium of claim 18 , wherein the glucose cost weight coefficient has a value of a baseline insulin cost weight coefficient multiplied by a value indicative of a ratio of a baseline value representative of insulin needs to a custom value representative of insulin needs of the user.
26. The non-transitory computer-readable storage medium of claim 18 , wherein the glucose cost weight coefficient has a value of a baseline insulin cost weight coefficient multiplied by a value indicative of a ratio of a baseline value representative of insulin needs to a custom value representative of insulin needs of the user.
27. The non-transitory computer-readable storage medium of claim 18 , further storing instruction for enforcing bounds on a parameter used in determining at least one of the glucose cost weight coefficient or the insulin cost weight coefficient that is customized for the user.
28. The non-transitory computer-readable storage medium of claim 27 , wherein the parameter is a value indicative of the insulin needs of the user.
29. The non-transitory computer-readable storage medium of claim 18 , wherein the determining of at least one of the glucose cost weight coefficient or the insulin cost weight coefficient is based on at least one of a correction factor for insulin sensitivity for the user, an insulin to carbohydrate ratio for the user or a basal insulin level for the user.
30. The non-transitory computer-readable storage medium of claim 18 , wherein at least one of the glucose cost weight coefficient and the insulin cost weight coefficient have values customized for the Total Daily Insulin (TDI) user.