a data driven approach to identify the pseudo-periodic behavior of blood glucose in diabetic patients

In this work we present a data driven approach to identify the pseudoperiodic dynamical structure in blood glucose in diabetic patients. in this work we usethe pre-breakfast, pre-lunch and pre-supper blood glucose measurements of 70 patientsover an extended period of time. we employ an algorithm that automatically learns theinner dynamics of the given data. this technique recursively discovers the underlyingstructure and is updated as new data emerges using radial basis functions. this methodis very fast and memory efficient. our modeling technique has done well to demonstratethe covering curves for the pseudo-periodic curves of blood glucose. the geometry ofthe covering curves identifies hyperglycaemia and hypoglycaemia in diabetic patients