Long-Term Glucose Forecasting Using a Physiological Model and Deconvolution of the Continuous Glucose Monitoring Signal

Abstract

(1) Objective: Blood glucose forecasting in type 1 diabetes (T1D) management is a maturing field with numerous algorithms being published and a few of them having reached the commercialisation stage. However, accurate long-term glucose predictions (e.g., >60 min), which are usually needed in applications such as precision insulin dosing (e.g., an artificial pancreas), still remain a challenge. In this paper, we present a novel glucose forecasting algorithm that is well-suited for long-term prediction horizons. The proposed algorithm is currently being used as the core component of a modular safety system for an insulin dose recommender developed within the EU-funded PEPPER (Patient Empowerment through Predictive PERsonalised decision support) project. (2) Methods: The proposed blood glucose forecasting algorithm is based on a compartmental composite model of glucose–insulin dynamics, which uses a deconvolution technique applied to the continuous glucose monitoring (CGM) signal for state estimation. In addition to commonly employed inputs by glucose forecasting methods (i.e., CGM data, insulin, carbohydrates), the proposed algorithm allows the optional input of meal absorption information to enhance prediction accuracy. Clinical data corresponding to 10 adult subjects with T1D were used for evaluation purposes. In addition, in silico data obtained with a modified version of the UVa-Padova simulator was used to further evaluate the impact of accounting for meal absorption information on prediction accuracy. Finally, a comparison with two well-established glucose forecasting algorithms, the autoregressive exogenous (ARX) model and the latent variable-based statistical (LVX) model, was carried out. (3) Results: For prediction horizons beyond 60 min, the performance of the proposed physiological model-based (PM) algorithm is superior to that of the LVX and ARX algorithms. When comparing the performance of PM against the secondly ranked method (ARX) on a 120 min prediction horizon, the percentage improvement on prediction accuracy measured with the root mean square error, A-region of error grid analysis (EGA), and hypoglycaemia prediction calculated by the Matthews correlation coefficient, was 18.8%, 17.9%, and 80.9%, respectively. Although showing a trend towards improvement, the addition of meal absorption information did not provide clinically significant improvements. (4) Conclusion: The proposed glucose forecasting algorithm is potentially well-suited for T1D management applications which require long-term glucose predictions.

Glucose Prediction under Variable-Length Time-Stamped Daily Events: A Seasonal Stochastic Local Modeling Framework

“…More recent in silico studies, in this case using the same extended UVA/Padova simulator as in this work, have been reported. In [ 27 ], a physiological model (PM) in combination with CGM signal deconvolution is presented for long-term glucose prediction. The model requires as inputs carbohydrate intake and insulin delivery, as opposed to our case.…”

Medical Assistant Mobile Application for Diabetes Control by Simulating a Compartmental Model

“…The literature review indicates that early studies of compartmental modelling of glucose/insulin interactions were proposed since the 1980s in [17][18][19]. On the other hand, recently, different research groups have reported important works related with the implementations of compartmental models to develop glucose level simulators, for example, in [20,21] are presented a glucose forecasting algorithm based on a compartmental composite model of glucose-insulin dynamics. In [22] is developed a dynamic model for type 1 diabetic patient that focuses on blood glucose regulation.…”

“…Similar results are obtained for 30 and 90 min of exercise duration. In general, there are many applications, however, long-term prediction of glucose still remains a challenge [20]. In this work, an app has been developed capable of predicting the behavior of glucose levels for several days (see Figure 5) or during a day with different exercise regimes (see Figure 6).…”

Identifiability of Control-Oriented Glucose-Insulin Linear Models: Review and Analysis

“…In [13] three minimal models were identified by minimizing the mean square error concerning the CGM measurements. In addition, a long-term glucose prediction algorithm based on a physiological model and a deconvolution technique using CGM data was presented in [21] by adding information about meal absorption to enhance prediction accuracy and using a constrained optimization technique that minimizes the mean absolute relative difference to identify 3 of the 10 parameters of the model.…”