CPC G16H 40/63 (2018.01) [G01N 33/6851 (2013.01); G01N 33/6893 (2013.01); G01N 2800/042 (2013.01)] | 5 Claims |
1. A system for predicting a blood glucose level of a pregnant individual, wherein the system comprises an operation module, and the operation model comprises a support vector regression model, and the system is configured to predict the blood glucose level at 1 hour after taking oral glucose under a fasting state by using concentrations of biomarkers in blood samples of the pregnant individual under the fasting state and the support vector regression model;
wherein the equation of the support vector regression model is:
where, Y is the predicted blood glucose level, i represents the ith biomarker, m represents a number of the biomarkers, Wi represents a weight of the ith biomarker, Ki represents a coefficient of the ith biomarker, and b is a constant;
wherein Ki is calculated by the following formula:
Ki=(γ·μ1·v1+coef)degree
where, γ, coef and degree are parameters to be adjusted, ui·vi is a linear model of independent variables, i is a linear coefficient of the ith biomarker, and vi is the concentration of the ith biomarker;
wherein m is 19;
wherein the biomarkers are glucose, 1,5-anhydroglucitol, 3-methyl-2-oxobutyrate, 3-hydroxybutyrate, 2-hydroxybutyrate, pantothenic acid, 3-methyl-2-oxovalerate, 4-methyl-2-oxovalerate, palmitoyl glycerol phosphatidylcholine (palmitoyl-GPC), palmitoylcarnitine, oleate, glycine, phenylalanine, serine, tyrosine, isoleucine, leucine, valine, and 2-aminoadipic acid.
|