On the identification of Pols in glucosimeter data

Diabetes is a chronic disease affecting a significant percentage (about 10%) of the population characterized by the presence of high level of glucose in the blood (and thus in tissues). Diabetes occurs when the pancreas does not produce enough insulin (type 1 diabetes) or when the body cannot effectively use the insulin it produces (type 2 diabetes). Monitoring the glucose blood level and assessing the insulin dosage is one of the main target for clinicians. To this end, identifying Pattern of Interests (PoIs) in glucose daily distribution measured by using glucosimeters, may help in personalizing insulin dosage.In this paper, we propose a new approach for analyzing glucosimeters acquired data exploiting time varying windows. The proposed approach is being used at University Magna Graecia Dismetabolic unit to study clinical data and to identify interesting patterns from glucose daily measured values.