(437d) Long-Term Stability of Inferential Blood Glucose Model Developed From Noninvasive Inputs

In this work we present the results validating the efficacy and accuracy of a block-oriented modeling approach in modeling blood glucose of a real type 2 diabetic (T2D) subject using non-invasive data from a real data and its ability to predict accurately over a period of almost two years. The selected 11 inputs for this study have nonlinear, highly interactive, and dynamic affects on blood glucose. A Wiener model used for this modeling consisted of linear dynamic blocks followed by a static nonlinear function. The full final Wiener model from this study consisted of the eleven variables and had 111 parameters.

The model was developed (trained) from 3 weeks of data collected on a T2D subject in September ?October 2006. The model was then tested on five days of data immediately after the training period (October 2006) and showed its excellent ability to model glucose over the test period using only non-invasive food and activity inputs. As a measure of evaluating performance, an average absolute error (AAE) (i.e., the average of the absolute values for the measured glucose minus modeled glucose) was used. The AAE in training was 12.0 mg/dL and 13.6 mg/dL in testing in 2006. While the ability of the model to predict well during testing in 2006, was itself noteworthy and indicative of its ability to give cause-and-effect, an additional testament to the model's capability is its ability to model the data collected on the same subject almost two years later in July 2008 at similar level of accuracy. (For comparison consider that the glucose meter AAE for replicated measurements was 15.3 mg/dL).

In addition to the full model (FM), we also evaluated a reduced model (RM) which did not consider the second-order effects from the inputs. We considered the performance of the RM also for the same data and were able to get comparable results as the FM in terms of accuracy and efficacy.