Background and Aims: The UVA/Padova Type 1 Diabetes Simulator (T1DMS) has proven to be very useful for the preclinical testing of control algorithms in adults, also thanks to the availability of data which allowed to refine and validate the in silico population. This facilitated several successful artificial pancreas inpatient and outpatient studies. Artificial pancreas studies in children are increasing and the T1DMS could play an important role in algorithmic development. The aim of this study is to assess and, if necessary, to update the in silico children included into the T1DMS. Method: The data set consisted of 13 T1DM prepubertals recruited among the centers of Milano, Napoli, Roma and Torino (Italy). Subjects wore a sensor-augmented insulin pump, and received a standardized breakfast (30g of carbohydrates [CHO]). The T1DMS model was identified from subcutaneous glucose sensor and insulin pump data using a Bayesian approach. Results: The model well fitted the data and provided precise estimates of model parameters. Our results suggest that meal glucose absorption is faster and insulin sensitivity lower, on average, than those included so far in the T1DMS (Figure 1). Based on these findings, a new children population was generated. The new T1DMS was then validated by comparing real vs simulated glucose curves in response to 30 g CHO meal: as a result, the new T1DMS well described the glucose variability observed in the data. Conclusion: The new in silico children population will enable a safer and effective preclinical testing of the closed-loop control in pediatric clinical trials.
IMPROVEMENT OF THE CHILDREN POPULATION INCLUDED INTO THE UVA/PADOVA TYPE 1 DIABETES SIMULATOR in 9th International Conference on Advanced Technologies & Treatment of Diabetes - Barcellona 2016
Iafusco D.Conceptualization
;
2016
Abstract
Background and Aims: The UVA/Padova Type 1 Diabetes Simulator (T1DMS) has proven to be very useful for the preclinical testing of control algorithms in adults, also thanks to the availability of data which allowed to refine and validate the in silico population. This facilitated several successful artificial pancreas inpatient and outpatient studies. Artificial pancreas studies in children are increasing and the T1DMS could play an important role in algorithmic development. The aim of this study is to assess and, if necessary, to update the in silico children included into the T1DMS. Method: The data set consisted of 13 T1DM prepubertals recruited among the centers of Milano, Napoli, Roma and Torino (Italy). Subjects wore a sensor-augmented insulin pump, and received a standardized breakfast (30g of carbohydrates [CHO]). The T1DMS model was identified from subcutaneous glucose sensor and insulin pump data using a Bayesian approach. Results: The model well fitted the data and provided precise estimates of model parameters. Our results suggest that meal glucose absorption is faster and insulin sensitivity lower, on average, than those included so far in the T1DMS (Figure 1). Based on these findings, a new children population was generated. The new T1DMS was then validated by comparing real vs simulated glucose curves in response to 30 g CHO meal: as a result, the new T1DMS well described the glucose variability observed in the data. Conclusion: The new in silico children population will enable a safer and effective preclinical testing of the closed-loop control in pediatric clinical trials.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
