The Research Least Squares Based on Ar Model of Glucose Prediction

Abstract:

Diabetes Clinical Research important task is to monitor and prevent the occurrence of high / low blood sugar events. Glucose monitoring system (Continous Glucose Monitoring System, CGMS) is used clinically in recent years gradually new blood glucose monitoring system, by measuring the concentration of glucose in interstitial fluid glucose fluctuations throughout the day to indirectly reflect the whole picture. It can be 30 minutes early to predict blood glucose levels as well as low blood sugar warning. Based on time-series modeling techniques, with blood glucose levels for non-stationary paper with improved self-regression model (AR) on blood glucose prediction. And the order of the model parameters were determined by the least squares method and adaptive AIC criteria. The results show that the algorithm can be based on time series modeling real-time and accurate display changes in blood glucose levels, while forecasting and early warning of low blood sugar glucose results showed good performance.

• Periodical: Advanced Materials Research (Volumes 971-973)
• Pages: 284-287
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.971-973.284
• Online since: June 2014
• Authors: Yan Nian Wang*, Bang An
• Keywords: AIC Criteria, Autoregressive AR Model, Blood Glucose Prediction, Glucose Monitoring System (CGMS), Least Squares Algorithm

References

[1] Albisser A M, Sakkal S, Wright C. Home blood glucose prediction: validation, safety, and efficacy testing in clinical diabetes[J]. Diabetes technology & therapeutics, 2005, 7(3): 487-496.

DOI: 10.1089/dia.2005.7.487

[2] Egi M, Bellomo R, Stachowski E, et al. Variability of blood glucose concentration and short-term mortality in critically ill patients[J]. Anesthesiology, 2006, 105(2): 244-252.

DOI: 10.1097/00000542-200608000-00006

[3] Turner R C, Holman R R, Cull C A, et al. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33)[J]. lancet, 1998, 352(9131): 837-853.

DOI: 10.1016/s0140-6736(98)07019-6

[4] Bayrak ES, Kamuran Turksoy, B. S, etal. Hypoglycemia Early Alarm Systems Based on Recursive Autoregressive Partial Least Squares Models[J]. Journal of Diabetes Science and Technology, 2013, 7(1): 206 – 214.

DOI: 10.1177/193229681300700126

[5] Gao F, Thorp J S, Pal A, et al. Dynamic state prediction based on auto-regressive (AR) model using PMU data[C]/Power and Energy Conference at Illinois (PECI), 2012 IEEE. IEEE, 2012: 1-5.

DOI: 10.1109/peci.2012.6184586

[6] Tatinati S, Veluvolu K C. Multi-step tremor prediction autoregressive (AR) model and Kalman filter (KF) for surgical robotic applications[C]/Power, Energy and Control (ICPEC), 2013 International Conference on. IEEE, 2013: 473-477.

DOI: 10.1109/icpec.2013.6527703

[7] Shimamura T, Yokose T. AR-model-based data extension to improve the Performance of MUSIC[C]/Intelligent Signal Processing and Communications Systems (ISPACS), 2013 International Symposium on. IEEE, 2013: 458-461.

DOI: 10.1109/ispacs.2013.6704593

[8] Sivananthan S, Naumova V, Man C D, et al. Assessment of blood glucose predictors: the prediction-error grid analysis[J]. Diabetes technology & therapeutics, 2011, 13(8): 787-796.

DOI: 10.1089/dia.2011.0033