Prediction Technology of Type Ⅱ Diabetes Based on Markov Chain

Diabetes is on the rise worldwide. Diabetes and its associated complications endanger the life of a large number of populations. Therefore, it is critical to predict and prevent the high incidence rate of diabetes. This paper proposes a method of long-term prediction of type Ⅱ diabetes incidence rate based on the Markov model. This method applies the naive Bayes arithmetic to calculate the probability vector of risk levels, and establishes the predicting model with selected property subsets of GLU, BMI, CHOL, TG, WAIST, SEX, DME and AGE. Compared with the predicted results obtained by using Archimedes model, the proposed method can predict long-term incidence rate of diabetes. The model recommended in this paper is proved to be simple and accurate for predicting the long-term incidence rate of type Ⅱ diabetes.