https://www.selleckchem.com/ 0%, and the false-negative-rate was 0.015%. Similar results were found even when only CGM samples below 70 were considered. The true-positive-hyperglycemia-prediction-rate was 61%. Conclusions State-of-the-art SML tools are effective in predicting the glucose level values of patients with type-1diabetes and notifying these patients of future hypoglycemic and hyperglycemic events, thus improving glycemic control. The algorithm can be used to improve the calculation of the basal insulin rate and bolus insulin, and suitable for a closed loop "artificial pancreas" system. The algorithm provides a personalized medical solution that can successfully identify the best-fit method for each patient. This article is protected by copyright. All rights reserved.Aims to build a tool to assess the management of inpatients with diabetes mellitus and to investigate its relationship, if any, with clinical outcomes. Materials and methods 678 patients from different settings, Internal Medicine (IMU, n = 255), General Surgery (GSU, n = 230) and Intensive Care (ICU, n = 193) Units, were enrolled. A work-flow of clinical care of diabetes was created according to guidelines. The workflow was divided in 5 different domains 1) initial assessment, 2) glucose monitoring, 3) medical therapy, 4) consultancies, 5) discharge. Each domain was assessed by a performance score (PS), computed as the sum of the scores achieved in a set of indicators of clinical appropriateness, management and patient empowerment. Appropriate glucose goals were included as intermediate phenotypes. Clinical outcomes included hypoglycemia, survival rate and clinical conditions at discharge. Results the total PS and those of initial assessment and glucose monitoring were significantly lower in GSU respect to IMU and ICU (P less then 0.0001). The glucose monitoring PS was associated with lower risk of hypoglycemia (OR 0.55; P less then 0.0001), whereas both the PSs of glucose monitoring and medic