CBSim covariant cohorts

Published: 7 May 2020| Version 1 | DOI: 10.17632/g6y6pyhzzb.1
Onofre Orozco-López,
Agustín Rodríguez-Herrero,
Carlos E. Castañeda,
Gema García-Sáez,
M. Elena Hernando


From the available reduced number of subjects in the CBSim, it is obtained the covariance of each parameter of every subject to establish a mathematical relationship. Then, two covariant cohorts of 256 subjects each one, are obtained with linear regression and there are shared in the "Covariant_Cohorts" file. These covariant cohorts are: 1.- Covariant cohort generated using the parameters published in Chassin's Thesis (Chassin, 2005). 2.- Covariant cohort generated using the parameters published by Hovorka (Hovorka et al., 2004). Cohorts are clustered to avoid the overlapping in the glucose-insulin dynamics and are used to compare their qualitative and quantitative behavior in the normoglycemic range.



Type 1 Diabetes