Trend Estimation of Blood Glucose Level Fluctuations Based on Data Mining

Abstract

We have fabricated calorie-calculating software that calculates and records the total calorific food intake by choosing a meal menu selected using a computer mouse. The purpose of this software was to simplify data collection throughout a person's normal life, even if they were inexperienced computer operators. Three portable commercial devices have also been prepared a blood glucose monitor, a metabolic rate monitor and a mobile-computer, and linked into the calorie-calculating software. Time-course changes of the blood glucose level, metabolic rate and food intake were measured using these devices during a 3 month period. Based on the data collected in this study we could predict blood glucose levels of the next morning (FBG) by modeling using data mining. Although a large error rate was found for predicting the absolute value, conditions could be found that improved the accuracy of the predicting trends in blood glucose level fluctuations by up to 90 %. However, in order to further improve the accuracy of estimation it was necessary to obtain further details about the patients' life style or to optimise the input variables that were dependent on each patient rather than collecting data over longer periods.

Keywords

• Data Mining
• Estimation
• Metabolic Rate
• Diabetic Patient
• Food Intake
• Blood Glucose