Page 311 - Glucose Monitoring Devices

318 CHAPTER 15 Automated closed-loop insulin delivery

(A)
1.4
1.2
Insulin (U/h) 0.8 1

Basal 0.6
0.4
0.2
0
1 2 3
Time(day)
(B)
1.4
1.2
Insulin (U/h) 0.8 1

Basal 0.6
0.4
0.2
0
1 2 3
Time(day)
Meanofclosed-loopinsulin ± Std.dev. of insulin
FIGURE 15.5
Closed-loop results in mGIPsim for the basal insulin infusion using (A) sAP and (B) mAP.

Future directions
Meals and physical activity pose significant challenges for glycemic control in peo-
ple with diabetes. Addressing these challenges in closed-loop automated insulin
delivery systems is a priority. Other disturbances and factors affecting euglycemia
include the sleep architecture and acute psychological stress. The characteristics
of sleep can affect the glycemic dynamics and insulin sensitivity during the subse-
quent day. Therefore recognizing the overnight sleep characteristics and adjusting
the insulin-dosing algorithms accordingly to counteract the dawn phenomenon or
the glycemic variations of the subsequent day will improve glycemic control. Psy-
chological stress is associated with the activation of certain counterregulatory hor-
mones, which cause an increase in the BGC. Detecting and responding to
episodes of stress is also a possible future direction for AP systems. As shown in
Fig. 15.8, these additional modules can form a comprehensive multivariable AP sys-
tem for safe and effective glucose control.