Charles W. Wei

Application of the Mixtures of Experts algorithm for signal processing in a noninvasive glucose monitoring system

Abstract The theory of Mixtures of Experts (MOE) [M. Jordan, R. Jacobs, Hierarchical mixtures of experts and the EM algorithm, Neural Computation 6 (2) (1994) 181–214; S.R. Waterhouse, D.J.C. MacKay, et al., in: D.S. Touretzky (Ed.), Bayesian methods for Mixtures of Experts, Advances in Neural Information Processing Systems, Vol. 8, MIT Press, Cambridge, MA, 1996, pp. 351–357; S.R. Waterhouse, Classification and regression using Mixtures of Experts, PhD Thesis, Cambridge University, Cambridge, 1997] was applied to the signal from a noninvasive glucose monitor for the purpose of converting raw signal data into blood glucose values. The MOE algorithm can be described as a generalized predictive method of data analysis. This method uses a superposition of multiple linear regressions, along with a switching algorithm, to predict outcomes. Any number of input/output variables are possible. The unknown coefficients in this method are determined by an optimization technique called the Expectation Maximization (EM) algorithm. The noninvasive GlucoWatch® biographer operation has been described [R.T. Kurnik, B. Berner, et al., Design and simulation of a reverse iontophoretic glucose monitoring device, J. Electrochem. Soc. 145 (12) (1998) 4119–4125]. Briefly, a small electrical current results in the transport of glucose beneath the skin to a hydrogel placed on the skin surface. Within the hydrogel, the glucose reacts with the enzyme glucose-oxidase to produce hydrogen peroxide. This hydrogen peroxide then diffuses to a platinum-based electrode, where it reacts to produce a current. The integral of this current (charge) over the sensing time is the signal used to measure extracted glucose. This process is repeated, yielding up to three measurements per hour. The data used for this analysis were obtained from diabetic subjects wearing the biographer over a 15-h period. The MOE inputs consisted of elapsed time, integrated current, blood glucose value at the calibration point, and a calibrated signal. The output was the value of blood glucose at each measurement. These training data were used to determine the unknown parameters in the MOE by the EM algorithm. Using a 3-h time point for calibrating the biographer, the mean absolute error (MAE) between the actual blood glucose and the blood glucose predicted with the MOE, was 14.4%.

Use of the GlucoWatch biographer in children and adolescents with diabetes.

Abstract: Objective:  This study was done to evaluate the accuracy and safety of measuring glucose with the GlucoWatch® biographer in children and adolescents with diabetes.

The effect of preapplication of corticosteroids on skin irritation and performance of the GlucoWatch G2 Biographer.

Skin irritation due to iontophoresis may limit the frequency of use of devices for drug delivery or transdermal extraction of analytes of clinical interest. This study examined whether preapplication of corticosteroid preparations could reduce skin irritation from iontophoresis used by the GlucoWatch G2 Biographer (Cygnus, Inc., Redwood City, CA) in monitoring interstitial glucose levels frequently and automatically. Numerous corticosteroid preparations were screened to identify formulations that did not interfere with adhesion of the Biographer to the skin or glucose sensing. Kenalog (Westwood-Squibb Pharmaceuticals, Inc., Buffalo, NY) (triamcinolone acetonide) and Cortizone-10 Quick Shot (Pfizer, Inc., New York, NY) (hydrocortisone) sprays were selected and, in a double-masked, randomized, controlled trial, were applied to the forearms of 66 subjects with diabetes and allowed to dry. Biographers were applied and worn for 15 h, and home blood glucose measurements were taken every 30 min to assess accuracy. Irritation was assessed periodically by trained observers and study subjects. Skin irritation was reduced by both corticosteroid sprays, with the fraction of subjects who experienced moderate irritation reduced by 57% and 43% for the Kenalog and Cortizone-10 Quick Shot sprays, respectively. The treatment effect persisted at the 1-week assessment. Preapplication of these preparations did not affect the clinical utility of interstitial glucose readings. Preapplication of Kenalog or Cortizone-10 Quick Shot sprays significantly reduced skin irritation due to iontophoresis, and did not interfere with glucose measurements. This approach may enable the minority of users who experience moderate to severe skin irritation to use the Biographer more frequently for diabetes management.