Abstract
The problem of real time prediction of blood glucose (BG) levels based on the readings from a continuous glucose monitoring (CGM) device is a problem of great importance in diabetes care, and therefore, has attracted a lot of research in recent years, especially based on machine learning. An accurate prediction with a 30, 60, or 90 min prediction horizon has the potential of saving millions of dollars in emergency care costs. In this paper, we treat the problem as one of function approximation, where the value of the BG level at time t+h (where h the prediction horizon) is considered to be an unknown function of d readings prior to the time t. This unknown function may be supported in particular on some unknown submanifold of the d-dimensional Euclidean space. While manifold learning is classically done in a semi-supervised setting, where the entire data has to be known in advance, we use recent ideas to achieve an accurate function approximation in a supervised setting; i.e., construct a model for the target function. We use the state-of-the-art clinically relevant PRED-EGA grid to evaluate our results, and demonstrate that for a real life dataset, our method performs better than a standard deep network, especially in hypoglycemic and hyperglycemic regimes. One noteworthy aspect of this work is that the training data and test data may come from different distributions.
Highlights
Diabetes is a major disease affecting humanity
The purpose of this paper is to use this new tool for blood glucose (BG) prediction based on Continuous glucose monitoring (CGM) readings which can be used on patients not in the same data sets, and for readings for the patients in the data set which are not included among the training data
We repeat the entire process described in Subsections 4.2–4.5 for a fixed number of trials, after which we report the average of the PRED-EGA grid placements, over all xj ∈ Q and over all trials, as the final evaluation
Summary
Diabetes is a major disease affecting humanity. According to the 2020 National Diabetes Statistics report [3], more than 34,000,000 people had diabetes in the United States alone in 2018, contributing to nearly 270,000 deaths and costing nearly 327 billion dollars in 2017. There is so far no cure for diabetes, and one of the important ingredients in keeping it in control is to monitor blood glucose levels regularly. Continuous glucose monitoring (CGM) devices are used increasingly for this purpose. These devices typically record blood sugar readings at 5 min intervals, resulting in a time series. Since the onset of hypoglycemic periods are often silent (without indicating symptoms), it can be very hard for a human to predict in advance. If these predictions are getting communicated to a hospital by some wearable communication device linked with the CGM device, a nurse could
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
