To search, Click below search items.


All Published Papers Search Service


Un- Interrupted Glucose Monitoring and Prediction by Using Smart Sensor System


Dr. Arshad Ali, Dr. Yazed Alsaawvy


Vol. 19  No. 4  pp. 211-218


The monitoring of real time Blood Glucose level in diabetes patient is a big problem to be addressed in near future to treat patients efficiently. There are some kind of the Continuous Glucose Monitoring (GCM) sensors available in market but very expensive to use for long term to monitor glucose level continuously. To overcome this problem a combination of the real time monitoring of glucose is used to predict the glucose concentration at a particular time. There are various prediction techniques are available and used to predict glucose concentration in blood. For the purpose of prediction of glucose level in this research, four different parameters are considered to predict glucose concentration by using Kriging prediction algorithm. First, the CGM data is calibrated accurately. Secondly, the CGM data is filtered to improve its signal-to-noise ratio (SNR). Thirdly, prediction of future glucose concentration is predicted by using appropriate modeling techniques. Lastly, the generation of alerts minimize the risk of detecting false and missing true events. Based on the study and experimentation Kriging based algorithm for prediction is performed very well as compared to other prediction technique with the lowest mean square error.


Continue Glucose Monitoring, Blood Glucose, Vector Support Engine, Root Mean Square Error, Weight Factor