Abstract
The artificial pancreas (AP) system is designed to regulate blood glucose in subjects with type 1 diabetes using a continuous glucose monitor informed controller that adjusts insulin infusion via an insulin pump. However, current AP developments are mainly hybrid closed-loop systems that include feed-forward actions triggered by the announcement of meals or exercise. The first step to fully closing the loop in the AP requires removing meal announcement, which is currently the most effective way to alleviate postprandial hyperglycemia due to the delay in insulin action. Here, a novel approach to meal detection in the AP is presented using a sliding window and computing the normalized cross-covariance between measured glucose and the forward difference of a disturbance term, estimated from an augmented minimal model using an Unscented Kalman Filter. Three different tunings were applied to the same meal detection algorithm: (1) a high sensitivity tuning, (2) a trade-off tuning that has a high amount of meals detected and a low amount of false positives (FP), and (3) a low FP tuning. For the three tunings sensitivities 99 ± 2%, 93 ± 5%, and 47 ± 12% were achieved, respectively. A sensitivity analysis was also performed and found that higher carbohydrate quantities and faster rates of glucose appearance result in favorable meal detection outcomes.
Highlights
Closed-loop systems intended for blood glucose (BG) control in people with type 1 diabetes (T1D), i.e. artificial pancreas (AP) systems, are still dependent on feed-forward actions such as meal announcements to achieve effective control
Three different tunings of the same detection algorithm were analyzed, one had the highest sensitivity, the second had a trade-off between the number of true positive (TP) and false positive (FP), and the third had the lowest number of FP
Studied nine in clinic patients over 32 hours with an average meal size of 44 ± 9.4 g. This meal detection algorithm has both a high sensitivity of 97 ± 6% and low ∆ glucose of 16 ± 9.4 mg/dL and, out of 63 meals and snacks, there was only one FP and two false negative (FN). Their results outperform the results found in this study, the duration of this study was not long enough to truly reflect meal detection capabilities
Summary
Closed-loop systems intended for blood glucose (BG) control in people with type 1 diabetes (T1D), i.e. artificial pancreas (AP) systems, are still dependent on feed-forward actions such as meal announcements to achieve effective control. Many studies have reported that a high number of missed meal boluses occur, especially in adolescents during insulin pump therapy. Several studies have reported a link between glycated hemoglobin (HbA1c) levels and missed meal boluses [1,2,3]. Showing an average increase in HbA1c of 4 mmol/mol (0.3%) during a 2-week period due to missed meal boluses [1]. These findings have been further compounded by Patton et al [4] who found a higher correlation between poor HbA1c levels and missed meal boluses compared to frequency of daily BG measures. It is predicted that this missed-meal-bolus behavior will carry over to AP therapy and, a means to reduce poor outcomes due to unannounced meals must be implemented
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
