A New Method for Estimating Diagnostic Parameters in the Dynamics Model of Modified Glucose-Insulin Homeostasis from the Oral Glucose Tolerance Test Using a Gravitational Search Algorithm

Monitoring the concentration of plasma glucose is very important to diagnose individuals with diabetes early so that the treatment of diabetes can be done early as well. The earlier diagnosis of the disease allows for better control. In general, the diagnostic determination of subjects with prediabetes or diabetes is based on the criteria of the basal threshold of measured plasma glucose statically. However, using the development of alternative diagnostic methods, the dynamic model of glucose-insulin homeostasis has become an alternative model to be used as a diagnostic model. In this study, the diagnostic model should be modified by adding the function of oral glucose absorption to the change rate of plasma glucose concentration. The modified model will adapt to the subject's diagnostic determination based on the oral glucose tolerance test (OGTT), and then the estimation of the diagnostic parameters of the model that fits the OGTT experimental data can be obtained using a modified gravitational search algorithm (GSA). The modified GSA was used to estimate the model parameters to diagnose subjects with characteristics of subjects with normal glucose tolerance (NGT), impaired glucose tolerance (IGT) or prediabetes, and type 2 diabetes mellitus (T2DM). The suitability of the simulation results and the OGTT experimental data is shown by the suitability of the correlation curve results which are assessed based on the coefficient of determination (R-2). If the value of R-2 is above 0.90, it means that the simulation results are said to have a good match with the experimental data. Two parameters that are commonly used as a diagnostic determination are glucose effectiveness which indicates the ability to decrease plasma glucose concentration by itself and insulin sensitivity which measures the effect of interstitial plasma insulin response when plasma glucose concentration increases. In addition, metabolic disorders caused by insulin resistance in diabetes will also be characterized by an increase in plasma glucose. Therefore, the modified model will also be used to diagnose insulin resistance based on OGTT. This diagnostic study shows good compatibility because it is evident that the parameter results obtained from the modified model are within the criteria range of subjects with NGT, IGT, and T2DM based on the literature.

Automated summary

Monitoring plasma glucose concentration is crucial for early diagnosis of diabetes, leading to better disease control. In addition to static glucose measurement, a dynamic glucose-insulin homeostasis model can be used as an alternative diagnostic model.