Towards blood glucose and insulin estimation using off-line data and missing information

Abstract

Type 1 diabetes mellitus is an autoimmune disease, in which the beta cells of the pacreas are destroyed, which are responsible for provomer the production and release of insulin, which makes the transport, absorption and storage of glucose impossible. This can lead to different more serious diseases. In order to find a treatment for type diabetes mellitus, insulin infusion therapies and advanced control techniques based on the monitoring of glucose levels have been developed. Monitoring blood glucose concentration in patients with type 1 diabetes mellitus is the first step to exert a control action that regulates glucose homeostasis, said monitoring can be carried out periodically and / or sporadically. However, physical sensors for continuous monitoring may fail and present data loss, which could cause a poor predictive diagnosis and thus generate erroneous control actions that lead the patient to episodes of hyperglycemia or hypoglycemia. Therefore, a mobile horizon estimator (MHE) is designed to reconstruct the glucose concentration taking into account the periodic measurements of the continuous glucose monitor (CGM) and the sporadic measurements of the glucome-terse Both measurements will improve the quality of the measured information and adjust the state estimate when the information is lost.