Fasting Blood Glucose Change Prediction Model Based on Medical Examination Data and Data Mining Techniques

Fasting blood glucose (FBG) is an important indicator for human's health. Prediction for FBG is meaningful for finding and healing diseases, especially for diabetes mellitus. Based on four years' historical medical examination data, a prediction model of coming year's FBG is presented using traditional data mining techniques with a novel algorithm to estimate the FBG change probability and a proposed feature selection algorithm, which combines the feature importance scores of ensemble learning and Sequential Backward Selection (SBS) algorithm to select an optimal feature subset. Experimental data are collected from a medical examination database containing 108,386 users, in which 7,136 people have four years' records. Compared with traditional support vector machine (SVM) and random forest, experimental results demonstrate that the feature selection algorithm can improve the performance of both SVM and random forest. Also the proposed method to estimate the probability of the FBG change works promisingly for giving an intuitive description of predictive result.