Development and validation of a predictive model of acute glucose response to exercise in individuals with type 2 diabetes

A simple modeling framework for prediction in the human glucose-insulin system

Forecasting blood glucose (BG) levels with routinely collected data is useful for glycemic management. BG dynamics are nonlinear, complex, and nonstationary, which can be represented by nonlinear models. However, the sparsity of routinely collected data creates parameter identifiability issues when high-fidelity complex models are used, thereby resulting in inaccurate forecasts. One can use models with reduced physiological fidelity for robust and accurate parameter estimation and forecasting with sparse data. For this purpose, we approximate the nonlinear dynamics of BG regulation by a linear stochastic differential equation: we develop a linear stochastic model, which can be specialized to different settings: type 2 diabetes mellitus (T2DM) and intensive care unit (ICU), with different choices of appropriate model functions. The model includes deterministic terms quantifying glucose removal from the bloodstream through the glycemic regulation system and representing the effect of nutrition and externally delivered insulin. The stochastic term encapsulates the BG oscillations. The model output is in the form of an expected value accompanied by a band around this value. The model parameters are estimated patient-specifically, leading to personalized models. The forecasts consist of values for BG mean and variation, quantifying possible high and low BG levels. Such predictions have potential use for glycemic management as part of control systems. We present experimental results on parameter estimation and forecasting in T2DM and ICU settings. We compare the model's predictive capability with two different nonlinear models built for T2DM and ICU contexts to have a sense of the level of prediction achieved by this model.

A qualitative study to explore the acceptability and usefulness of personalized biofeedback to motivate physical activity in cancer survivors

Objective

Many cancer survivors do not meet recommended levels of exercise, despite the benefits physical activity offers. This study aimed to understand experiences of insufficiently active overweight/obese breast or colorectal cancer survivors, in efforts to (1) examine regular physical activity barriers, and (2) determine perceptions and acceptability of a remotely delivered physical activity intervention utilizing wearable sensors and personalized feedback messages.

Methods

In-person and virtual small group interviews were conducted engaging overweight/obese cancer survivors (n = 16, 94% female, 94% breast cancer survivors) in discussions resulting in 314 pages of transcribed data analyzed by multiple coders.

Results

All participants expressed needing to increase physical activity, identifying lack of motivation centering on survivorship experiences and symptom management as the most salient barrier. They indicated familiarity with activity trackers (i.e., Fitbit) and expressed interest in biosensors (i.e., continuous glucose monitors [CGMs]) as CGMs show biological metrics in real-time. Participants reported (1) personalized feedback messages can improve motivation and accountability; (2) CGM acceptability is high given survivors’ medical history; and (3) glucose data is a relevant health indicator and they appreciated integrated messages (between Fitbit and CGM) in demonstrating how behaviors immediately affect one's body.

Conclusions

This study supports the use of wearable biosensors and m-health interventions to promote physical activity in cancer survivors. Glucose-based biofeedback provides relevant and motivating information for cancer survivors regarding their daily activity levels by demonstrating the immediate effects of physical activity. Integrating biofeedback into physical activity interventions could be an effective behavioral change strategy to promote a healthy lifestyle in cancer survivors.

Development of a Deep Learning Model for Dynamic Forecasting of Blood Glucose Level for Type 2 Diabetes Mellitus: Secondary Analysis of a Randomized Controlled Trial

BACKGROUND

Type 2 diabetes mellitus (T2DM) is a major public health burden. Self-management of diabetes including maintaining a healthy lifestyle is essential for glycemic control and to prevent diabetes complications. Mobile-based health data can play an important role in the forecasting of blood glucose levels for lifestyle management and control of T2DM.

OBJECTIVE

The objective of this work was to dynamically forecast daily glucose levels in patients with T2DM based on their daily mobile health lifestyle data including diet, physical activity, weight, and glucose level from the day before.

METHODS

We used data from 10 T2DM patients who were overweight or obese in a behavioral lifestyle intervention using mobile tools for daily monitoring of diet, physical activity, weight, and blood glucose over 6 months. We developed a deep learning model based on long short-term memory-based recurrent neural networks to forecast the next-day glucose levels in individual patients. The neural network used several layers of computational nodes to model how mobile health data (food intake including consumed calories, fat, and carbohydrates; exercise; and weight) were progressing from one day to another from noisy data.

RESULTS

The model was validated based on a data set of 10 patients who had been monitored daily for over 6 months. The proposed deep learning model demonstrated considerable accuracy in predicting the next day glucose level based on Clark Error Grid and ±10% range of the actual values.

CONCLUSIONS

Using machine learning methodologies may leverage mobile health lifestyle data to develop effective individualized prediction plans for T2DM management. However, predicting future glucose levels is challenging as glucose level is determined by multiple factors. Future study with more rigorous study design is warranted to better predict future glucose levels for T2DM management.

An Interactive Simulation to Change Outcome Expectancies and Intentions in Adults With Type 2 Diabetes: Within-Subjects Experiment

Background

Computerized simulations are underutilized to educate or motivate patients with chronic disease.

Objective

The aim of this study was to test the efficacy of an interactive, personalized simulation that demonstrates the acute effect of physical activity on blood glucose. Our goal was to test its effects on physical activity-related outcome expectancies and behavioral intentions among adults with type 2 diabetes mellitus (T2DM).

Methods

In this within-subjects experiment, potential participants were emailed a link to the study website and directed through 7 tasks: (1) consent; (2) demographics, baseline intentions, and self-reported walking; (3) orientation to the diurnal glucose curve; (4) baseline outcome expectancy, measured by a novel drawing task in which participants use their mouse to draw the expected difference in the diurnal glucose curve if they had walked; (5) interactive simulation; (6) postsimulation outcome expectancy measured by a second drawing task; and (7) final measures of intentions and impressions of the website. To test our primary hypothesis that participants’ outcome expectancies regarding walking would shift toward the outcome presented in the interactive simulation, we used a paired t test to compare the difference of differences between the change in area under the curve in the simulation and participants’ two drawings. To test whether intentions to walk increased, we used paired t tests. To assess the intervention’s usability, we collected both quantitative and qualitative data on participants’ perceptions of the drawing tasks and simulation.

Results

A total of 2019 individuals visited the website and 1335 (566 males, 765 females, and 4 others) provided complete data. Participants were largely late middle-aged (mean=59.8 years; standard deviation=10.5), female 56.55% (755/1335), Caucasian 77.45% (1034/1335), lower income 64.04% (855/1335) t₁₃₃₄=3.4, P ≤.001). Our second hypothesis, that participants’ intentions to walk in the coming week would increase, was also supported; general intention (mean difference=0.31/7, t₁₀₀₁=10.8, P<.001) and minutes of walking last week versus planned for coming week (mean difference=33.5 min, t₁₃₃₄=13.2, P<.001) both increased. Finally, an examination of qualitative feedback and drawing task data suggested that some participants had difficulty understanding the website. This led to a post-hoc subset analysis. In this analysis, effects for our hypothesis regarding outcome expectancies were markedly stronger, suggesting that further work is needed to determine moderators of the efficacy of this simulation.

Conclusions

A novel interactive simulation is efficacious in changing the outcome expectancies and behavioral intentions of adults with T2DM. We discuss applications of our results to the design of mobile health (mHealth) interventions.

Personalized glucose forecasting for type 2 diabetes using data assimilation

Type 2 diabetes leads to premature death and reduced quality of life for 8% of Americans. Nutrition management is critical to maintaining glycemic control, yet it is difficult to achieve due to the high individual differences in glycemic response to nutrition. Anticipating glycemic impact of different meals can be challenging not only for individuals with diabetes, but also for expert diabetes educators. Personalized computational models that can accurately forecast an impact of a given meal on an individual's blood glucose levels can serve as the engine for a new generation of decision support tools for individuals with diabetes. However, to be useful in practice, these computational engines need to generate accurate forecasts based on limited datasets consistent with typical self-monitoring practices of individuals with type 2 diabetes. This paper uses three forecasting machines: (i) data assimilation, a technique borrowed from atmospheric physics and engineering that uses Bayesian modeling to infuse data with human knowledge represented in a mechanistic model, to generate real-time, personalized, adaptable glucose forecasts; (ii) model averaging of data assimilation output; and (iii) dynamical Gaussian process model regression. The proposed data assimilation machine, the primary focus of the paper, uses a modified dual unscented Kalman filter to estimate states and parameters, personalizing the mechanistic models. Model selection is used to make a personalized model selection for the individual and their measurement characteristics. The data assimilation forecasts are empirically evaluated against actual postprandial glucose measurements captured by individuals with type 2 diabetes, and against predictions generated by experienced diabetes educators after reviewing a set of historical nutritional records and glucose measurements for the same individual. The evaluation suggests that the data assimilation forecasts compare well with specific glucose measurements and match or exceed in accuracy expert forecasts. We conclude by examining ways to present predictions as forecast-derived range quantities and evaluate the comparative advantages of these ranges.