Intensive insulin therapy has an inherent risk of hypoglycemia that can lead to loss of consciousness, cardiac arrhythmia, seizure, and death (“dead-in-bed syndrome”). This risk of hypoglycemia is a major concern for patients, families, and physicians. The need for an automated system that can alert in the event of severe hypoglycemia is evident. In engineering systems, where there is a risk of malfunction of the primary control system, alert and safety mechanisms are implemented in layers of protection. This concept has been adopted in the proposed system that integrates a hypoglycemia prediction algorithm with a global position system (GPS) locator and short message service such that the current glucose value with the rate of change (ROC) and the location of the subject can be communicated to a predefined list. Furthermore, if the system is linked to the insulin pump, it can suspend the pump or decrease the basal insulin infusion rate to prevent the pending event. The system was evaluated on clinical datasets of glucose tracings from the DexCom Seven® system. Glucose tracings were analyzed for hypoglycemia events and then a text message was broadcast to a predefined list of people who were notified with the glucose value, ROC, GPS coordinates, and a Google map of the location. In addition to providing a safety layer to a future artificial pancreas, this system also can be easily implemented in current continuous glucose monitors to help provide information and alerts to people with diabetes.
A Variable Prediction Horizon Self-tuning Method for Nonlinear Model Predictive Speed Control on PMSM Rotor Position System
