Description:
Background
Continuous glucose sensors (CGSs) offer the potential to markedly change our understanding of glucose homeostasis in health and disease and to provide the missing information required to achieve near-normoglycemia among persons with both Type 1 and Type 2 diabetes.
Despite their having appeared on the scene less than five years ago, CGSs already have been shown to be associated with short-term reductions in glucose variability, time spent in nocturnal hypoglycemia, time spent in hyperglycemia and lower glycosylated hemoglobin values. Their potential to assist patients and their families in day-to-day decision-making, to warn of impending hypoglycemia and to reduce the fear of its occurrence has yet to be fully appreciated. The advent of CGS has stimulated the diabetes research community to re-examine the feasibility of developing a closed-loop “artificial pancreas.”
Unfortunately, much of the data regarding CGS accuracy is presented according to criteria developed decades ago for assessing the accuracy of home blood glucose monitors providing episodic self-monitoring readings and thus do not include assessments designed to evaluate “continuous” time-dependent information (in particular rate and direction of glucose change), which is unique to CGSs.
About the Invention
In light of the fact that sensor accuracy is today still substantially inferior to the accuracy of self-monitoring devices that use blood from finger sticks for glucose determination, researchers at the University of Virginia have developed an algorithm that focuses, for the first time, on the signal from the insulin pump as a source of information for improvement of the accuracy of the CGS that is coupled with the insulin pump.
The method consists of three steps:
- Glucose state estimation from CGS readings, insulin delivery data from the insulin pump and Kalman Filter (KF) methodology to estimate the glycemic state of the person;
- Blood glucose prediction and projection of glucose fluctuations a few minutes ahead; and
- Weighting of the glucose state estimate against the sensor readings and calculation of a weighted blood glucose estimate.
As insulin delivery data becomes available to the sensor processor, the effect of using insulin information to enhance sensor accuracy is most prominent at low blood glucose levels (e.g., in the hypoglycemic range), which is critical for any treatment.
