A Deep Learning Approach to Diabetic Blood Glucose Prediction

We consider the question of 30-minute prediction of blood glucose levels measured by continuous glucose monitoring devices, using clinical data. While most studies of this nature deal with one patient at a time, we take a certain percentage of patients in the data set as training data, and test on the remainder of the patients; i.e., the machine need not re-calibrate on the new patients in the data set. We demonstrate how deep learning can outperform shallow networks in this example. One novelty is to demonstrate how a parsimonious deep representation can be constructed using domain knowledge.

[1]  T. Poggio,et al.  The Mathematics of Learning: Dealing with Data , 2005, 2005 International Conference on Neural Networks and Brain.

[2]  L. Quinn,et al.  Estimation of future glucose concentrations with subject-specific recursive linear models. , 2009, Diabetes technology & therapeutics.

[3]  D. Cox,et al.  Evaluating Clinical Accuracy of Systems for Self-Monitoring of Blood Glucose , 1987, Diabetes Care.

[4]  Hrushikesh Narhar Mhaskar,et al.  Filtered Legendre expansion method for numerical differentiation at the boundary point with application to blood glucose predictions , 2013, Appl. Math. Comput..

[5]  Ohad Shamir,et al.  Depth Separation in ReLU Networks for Approximating Smooth Non-Linear Functions , 2016, ArXiv.

[6]  T. Poggio,et al.  Deep vs. shallow networks : An approximation theory perspective , 2016, ArXiv.

[7]  Razvan Pascanu,et al.  On the Number of Linear Regions of Deep Neural Networks , 2014, NIPS.

[8]  Sergei V. Pereverzev,et al.  Legendre polynomials as a recommended basis for numerical differentiation in the presence of stochastic white noise , 2013 .

[9]  Sergei V. Pereverzyev,et al.  A meta-learning approach to the regularized learning - Case study: Blood glucose prediction , 2012, Neural Networks.

[10]  Sergei V. Pereverzyev,et al.  Adaptive parameter choice for one-sided finite difference schemes and its application in diabetes technology , 2012, J. Complex..

[11]  S. Pereverzyev,et al.  Extrapolation in variable RKHSs with application to the blood glucose reading , 2011 .

[12]  A. Savitzky,et al.  Smoothing and Differentiation of Data by Simplified Least Squares Procedures. , 1964 .

[13]  A. Singer.  From graph to manifold Laplacian: The convergence rate , 2006 .

[14]  H. N. Mhaskar,et al.  Eignets for function approximation on manifolds , 2009, ArXiv.

[15]  K. Shadan,et al.  Available online: , 2012 .

[16]  C. Cobelli,et al.  Assessment of blood glucose predictors: the prediction-error grid analysis. , 2011, Diabetes technology & therapeutics.

[17]  Mikhail Belkin,et al.  Towards a theoretical foundation for Laplacian-based manifold methods , 2005, J. Comput. Syst. Sci..

[18]  Scott M. Pappada,et al.  Neural network-based real-time prediction of glucose in patients with insulin-dependent diabetes. , 2011, Diabetes technology & therapeutics.

[19]  Brent D. Cameron,et al.  Development of a Neural Network for Prediction of Glucose Concentration in Type 1 Diabetes Patients , 2008, Journal of diabetes science and technology.

[20]  Linda G. Snetselaar.  Nutrition Counseling Skills for the Nutrition Care Process , 2008 .

[21]  W. Tamborlane,et al.  Limitations of conventional methods of self-monitoring of blood glucose: lessons learned from 3 days of continuous glucose sensing in pediatric patients with type 1 diabetes. , 2001, Diabetes care.

[22]  Hrushikesh Narhar Mhaskar,et al.  Locally Learning Biomedical Data Using Diffusion Frames , 2012, J. Comput. Biol..

[23]  Matus Telgarsky,et al.  Representation Benefits of Deep Feedforward Networks , 2015, ArXiv.

[24]  Giovanni Sparacino,et al.  Glucose Concentration can be Predicted Ahead in Time From Continuous Glucose Monitoring Sensor Time-Series , 2007, IEEE Transactions on Biomedical Engineering.

[25]  Srinivasan Rajaraman,et al.  Predictive Monitoring for Improved Management of Glucose Levels , 2007, Journal of diabetes science and technology.

[26]  Hrushikesh Narhar Mhaskar,et al.  Diffusion polynomial frames on metric measure spaces , 2008 .