Evaluating self-monitoring blood glucose strategies using a diabetic-patient software agent

We propose the method of calculating the cross-correlation between continuous blood glucose and interpolated values of blood glucose samples in order to evaluate blood glucose monitoring frequency in a diabetic-patient software-agent model. The patient software-agent model is a 24-hr circadian, self-aware, stochastic model of a diabetic patient´s blood glucose levels in a mobile agent environment. Monitoring frequency can vary from six times per day to as little as one time per week. Today, the cost of monitoring diabetes in the population is almost equal to the cost of its treatment. Due to the cost, discomfort and potential of infection related to blood glucose monitoring, human diabetic patients are looking for the optimal monitoring frequency which will provide good understanding of blood glucose levels but will keep blood sampling to a minimum. The proposed method quantitatively assesses various monitoring protocols regarding the monitoring frequency in nine predefined categories (health status and age) of patient agents.