A new model for the hepatic insulin secretion rate

The endocrine glands detect, produce and secrete hormones which through the blood stream affects other cells and thereby have a regulating behaviour on the entire endocrine system. Hormones are extremely important for the metabolism of a human being and impairments in the endocrine system may lead to severe diseases such as diabetes mellitus, osteoporosis, thyroiditis, cretinism or gigantismus. A thorough understanding of the endocrine system is clearly important in the medical treatment of endocrine diseases. The rate at which the endocrine glands secrete hormones is of particular interest but typically hormones undergo a large and variable extraction in other organs before and after reaching blood plasma, e.g. insulin is absorbed by the lever before reaching blood plasma. Consequently hormone secretion rates are not subject to direct measurement and thus inference becomes a challenging ill-posed inverse problem.

An important example within the endocrine system is the problem of reconstructing the pancreatic insulin secretion rate. A deeper understanding here may lead to a better understanding of the human glucose regulating system allowing for a detailed insight into the function of the pancreatic insulin-producing beta-cells, which can be used in early classification, prognosis and therapy of diabetes in human beings. Several approaches have been proposed based upon deconvolution of C-peptide data, however, the various models presented for assessing the insulin secretion rate are based upon C-peptide and insulin measurements only.

In the project a new model incorporating glucose levels is developed. This model is highly complex and leads to the combination of two ill-posed inverse problems. In the project sophisticated new statistical methodology for tackling this problem is considered possibly allowing for early detection of diabetes mellitus and improved treatment.