Back to results
US 11,439,754 B1Granted
Optimizing embedded formulations for drug delivery
AssigneeInsulet Corporation
InventorsJoon Bok Lee, Yibin Zheng, Jason O'Connor, Ashutosh Zade
Life of this patent
Estimating the term…
Abstract
Disclosed herein is a method for execution by a drug delivery device for determining an optimal dose of a liquid drug for current cycle of a medication delivery algorithm, the method utilizing a stepwise evaluation of a model and a cost function across a coarse search space consisting of coarse discrete quantities of the drug and a refined search space consisting of refined discrete quantities of the drug.
Claims20
A method of determining a dose of a drug for a current cycle of a medication delivery algorithm in an automated drug delivery system comprising:
- defining a first coarse search space by coarsely delineating a range of possible doses of the drug into coarse discrete quantities;
- evaluating a model and a cost function for the current cycle and a predetermined number of future cycles for each coarse discrete quantity in the first coarse search space;
- defining a refined search space by delineating, into refined discrete quantities, a range of doses of the drug centered around the coarse discrete quantity having the lowest cost, the refined search space having a finer delineation than the first coarse search space;
- evaluating the model and the cost function for the current cycle and the predetermined number of future cycles, wherein the current cycle is evaluated over the refined search space and the predetermined number of future cycles are evaluated over a second coarse search space;
- and
- selecting the refined discrete quantity having the lowest cost as the recommended dose of the drug.
3 more elements
A system comprising:
- a processor;
- software implementing a medication delivery algorithm, executed by the processor, the software determining a recommended dose of a drug for a current cycle of the medication delivery algorithm by performing the functions of:
- defining a first coarse search space by coarsely delineating a range of possible doses of the drug into coarse discrete quantities;
- evaluating a model and a cost function for the current cycle and a predetermined number of future cycles for each coarse discrete quantity in the first coarse search space;
- defining a refined search space by delineating, into refined discrete quantities, a range of doses of the drug centered around the coarse discrete quantity having the lowest cost, the refined search space having a finer delineation than the first coarse search space;
- evaluating the model and the cost function for the current cycle and the predetermined number of future cycles, wherein the current cycle is evaluated over the refined search space and the predetermined number of future cycles are evaluated over the second coarse search space;
- and
- selecting the refined discrete quantity having the lowest cost as the recommended dose of the drug.
5 more elements
Dependencies are inferred from the claim text.