GoCARB in the Context of an Artificial Pancreas

Accuracy and Feasibility of Using a Smartphone Application for Carbohydrate Counting Versus Traditional Carbohydrate Counting for Adults With Insulin-Treated Diabetes

BACKGROUND

Patients with insulin-treated diabetes struggle with performing accurate carbohydrate counting for proper blood glucose control. Little is known about the comparative accuracy and feasibility of carbohydrate counting methods.

PURPOSE

The purpose of this study was to determine whether carbohydrate counting using a smartphone application is more accurate and feasible than a traditional method.

THEORETICAL/CONCEPTUAL FRAMEWORK

Based on a conceptual model derived from the Technology Acceptance Model, feasibility was defined as usefulness, ease of use, and behavioral intention to use each method.

METHODS

A standardized meal was presented to 20 adults with insulin-treated diabetes who counted carbohydrates using traditional and smartphone methods. Accuracy was measured by comparing carbohydrate counting estimates with the standardized meal values. Perceived feasibility (usefulness, ease of use, behavioral intention) was measured using rating forms derived from the Technology Acceptance Model.

RESULTS

The number of training and estimation minutes were significantly higher for the traditional method than the smartphone method (Z = -3.83, P < .05; Z = -2.30, P < .05). The traditional method took an additional 1.4 minutes for estimation and 12.5 minutes for training. There were no significant differences in accuracy between traditional and smartphone methods for carbohydrate counting (Wilcoxon signed-rank test, Z = -1.10, P = .28). There were no significant differences between traditional and smartphone methods for feasibility (usefulness, Z = -.10, P = .95; ease of use, Z = -.36, P = .72; or behavioral intention, Z = -.94, P = .35).

CONCLUSION

While both traditional and smartphone methods were found to be similar in terms of accuracy and feasibility, the smartphone method took less time for training and for carbohydrate estimation.

A Randomized Crossover Trial to Compare Automated Insulin Delivery (the Artificial Pancreas) With Carbohydrate Counting or Simplified Qualitative Meal-Size Estimation in Type 1 Diabetes

OBJECTIVE

Qualitative meal-size estimation has been proposed instead of quantitative carbohydrate (CHO) counting with automated insulin delivery. We aimed to assess the noninferiority of qualitative meal-size estimation strategy.

RESEARCH DESIGN AND METHODS

We conducted a two-center, randomized, crossover, noninferiority trial to compare 3 weeks of automated insulin delivery with 1) CHO counting and 2) qualitative meal-size estimation in adults with type 1 diabetes. Qualitative meal-size estimation categories were low, medium, high, or very high CHO and were defined as <30 g, 30-60 g, 60-90 g, and >90 g CHO, respectively. Prandial insulin boluses were calculated as the individualized insulin to CHO ratios multiplied by 15, 35, 65, and 95, respectively. Closed-loop algorithms were otherwise identical in the two arms. The primary outcome was time in range 3.9-10.0 mmol/L, with a predefined noninferiority margin of 4%.

RESULTS

A total of 30 participants completed the study (n = 20 women; age 44 (SD 17) years; A1C 7.4% [0.7%]). The mean time in the 3.9-10.0 mmol/L range was 74.1% (10.0%) with CHO counting and 70.5% (11.2%) with qualitative meal-size estimation; mean difference was -3.6% (8.3%; noninferiority P = 0.78). Frequencies of times at <3.9 mmol/L and <3.0 mmol/L were low (<1.6% and <0.2%) in both arms. Automated basal insulin delivery was higher in the qualitative meal-size estimation arm (34.6 vs. 32.6 units/day; P = 0.003).

CONCLUSIONS

Though the qualitative meal-size estimation method achieved a high time in range and low time in hypoglycemia, noninferiority was not confirmed.

An automated all-in-one system for carbohydrate tracking, glucose monitoring, and insulin delivery

Glycemic control through titration of insulin dosing remains the mainstay of diabetes mellitus treatment. Insulin therapy is generally divided into dosing with long- and short-acting insulin, where long-acting insulin provides basal coverage and short-acting insulin supports glycemic excursions associated with eating. The dosing of short-acting insulin often involves several steps for the user including blood glucose measurement and integration of potential carbohydrate loads to inform safe and appropriate dosing. The significant burden placed on the user for blood glucose measurement and effective carbohydrate counting can manifest in substantial effects on adherence. Through the application of computer vision, we have developed a smartphone-based system that is able to detect the carbohydrate load of food by simply taking a single image of the food and converting that information into a required insulin dose by incorporating a blood glucose measurement. Moreover, we report the development of comprehensive all-in-one insulin delivery systems that streamline all operations that peripheral devices require for safe insulin administration, which in turn significantly reduces the complexity and time required for titration of insulin. The development of an autonomous system that supports maximum ease and accuracy of insulin dosing will transform our ability to more effectively support patients with diabetes.

Closed-Loop Insulin Delivery Systems: Past, Present, and Future Directions

Closed-loop (artificial pancreas) systems for automated insulin delivery have been likened to the holy grail of diabetes management. The first iterations of glucose-responsive insulin delivery were pioneered in the 1960s and 1970s, with the development of systems that used venous glucose measurements to dictate intravenous infusions of insulin and dextrose in order to maintain normoglycemia. Only recently have these bulky, bedside technologies progressed to miniaturized, wearable devices. These modern closed-loop systems use interstitial glucose sensing, subcutaneous insulin pumps, and increasingly sophisticated algorithms. As the number of commercially available hybrid closed-loop systems has grown, so too has the evidence supporting their efficacy. Future challenges in closed-loop technology include the development of fully closed-loop systems that do not require user input for meal announcements or carbohydrate counting. Another evolving avenue in research is the addition of glucagon to mitigate the risk of hypoglycemia and allow more aggressive insulin dosing.

Assessing Mealtime Macronutrient Content: Patient Perceptions Versus Expert Analyses via a Novel Phone App

Background: People with type 1 diabetes estimate meal carbohydrate content to accurately dose insulin, yet, protein and fat content of meals also influences postprandial glycemia. We examined accuracy of macronutrient content estimation via a novel phone app. Participant estimates were compared with expert nutrition analyses performed via the Remote Food Photography Method© (RFPM©). Methods: Data were collected through a novel phone app. Participants were asked to take photos of meals/snacks on the day of and day after scheduled exercise, enter carbohydrate estimates, and categorize meals as low, typical, or high protein and fat. Glycemia was measured via continuous glucose monitoring. Results: Participants (n = 48) were 15-68 years (34 ± 14 years); 40% were female. The phone app plus RFPM© analysis captured 88% ± 29% of participants' estimated total energy expenditure. The majority (70%) of both low-protein and low-fat meals were accurately classified. Only 22% of high-protein meals and 17% of high-fat meals were accurately classified. Forty-nine percent of meals with <30 g of carbohydrates were overestimated by an average of 25.7 ± 17.2 g. The majority (64%) of large carbohydrate meals (≥60 g) were underestimated by an average of 53.6 ± 33.8 g. Glycemic response to large carbohydrate meals was similar between participants who underestimated or overestimated carbohydrate content, suggesting that factors beyond carbohydrate counting may impact postprandial glycemic response. Conclusions: Accurate estimation of total macronutrients in meals could be leveraged to improve insulin decision support tools and closed loop insulin delivery systems; development of tools to improve macronutrient estimation skills should be considered.

An Insulin Bolus Advisor for Type 1 Diabetes Using Deep Reinforcement Learning

(1) Background: People living with type 1 diabetes (T1D) require self-management to maintain blood glucose (BG) levels in a therapeutic range through the delivery of exogenous insulin. However, due to the various variability, uncertainty and complex glucose dynamics, optimizing the doses of insulin delivery to minimize the risk of hyperglycemia and hypoglycemia is still an open problem. (2) Methods: In this work, we propose a novel insulin bolus advisor which uses deep reinforcement learning (DRL) and continuous glucose monitoring to optimize insulin dosing at mealtime. In particular, an actor-critic model based on deep deterministic policy gradient is designed to compute mealtime insulin doses. The proposed system architecture uses a two-step learning framework, in which a population model is first obtained and then personalized by subject-specific data. Prioritized memory replay is adopted to accelerate the training process in clinical practice. To validate the algorithm, we employ a customized version of the FDA-accepted UVA/Padova T1D simulator to perform in silico trials on 10 adult subjects and 10 adolescent subjects. (3) Results: Compared to a standard bolus calculator as the baseline, the DRL insulin bolus advisor significantly improved the average percentage time in target range (70-180 mg/dL) from 74.1%±8.4% to 80.9%±6.9% (p<0.01) and 54.9%±12.4% to 61.6%±14.1% (p<0.01) in the the adult and adolescent cohorts, respectively, while reducing hypoglycemia. (4) Conclusions: The proposed algorithm has the potential to improve mealtime bolus insulin delivery in people with T1D and is a feasible candidate for future clinical validation.

Artificial Intelligence for Diabetes Management and Decision Support: Literature Review

Background

Artificial intelligence methods in combination with the latest technologies, including medical devices, mobile computing, and sensor technologies, have the potential to enable the creation and delivery of better management services to deal with chronic diseases. One of the most lethal and prevalent chronic diseases is diabetes mellitus, which is characterized by dysfunction of glucose homeostasis.

Objective

The objective of this paper is to review recent efforts to use artificial intelligence techniques to assist in the management of diabetes, along with the associated challenges.

Methods

A review of the literature was conducted using PubMed and related bibliographic resources. Analyses of the literature from 2010 to 2018 yielded 1849 pertinent articles, of which we selected 141 for detailed review.

Results

We propose a functional taxonomy for diabetes management and artificial intelligence. Additionally, a detailed analysis of each subject category was performed using related key outcomes. This approach revealed that the experiments and studies reviewed yielded encouraging results.

Conclusions

We obtained evidence of an acceleration of research activity aimed at developing artificial intelligence-powered tools for prediction and prevention of complications associated with diabetes. Our results indicate that artificial intelligence methods are being progressively established as suitable for use in clinical daily practice, as well as for the self-management of diabetes. Consequently, these methods provide powerful tools for improving patients’ quality of life.

The challenges of achieving postprandial glucose control using closed-loop systems in patients with type 1 diabetes

For patients with type 1 diabetes, closed-loop delivery systems (CLS) combining an insulin pump, a glucose sensor and a dosing algorithm allowing a dynamic hormonal infusion have been shown to improve glucose control when compared with conventional therapy. Yet, reducing glucose excursion and simplification of prandial insulin doses remain a challenge. The objective of this literature review is to examine current meal-time strategies in the context of automated delivery systems in adults and children with type 1 diabetes. Current challenges and considerations for post-meal glucose control will also be discussed. Despite promising results with meal detection, the fully automated CLS has yet failed to provide comparable glucose control to CLS with carbohydrate-matched bolus in the post-meal period. The latter strategy has been efficient in controlling post-meal glucose using different algorithms and in various settings, but at the cost of a meal carbohydrate counting burden for patients. Further improvements in meal detection algorithms or simplified meal-priming boluses may represent interesting avenues. The greatest challenges remain in regards to the pharmacokinetic and dynamic profiles of available rapid insulins as well as sensor accuracy and lag-time. New and upcoming faster acting insulins could provide important benefits. Multi-hormone CLS (eg, dual-hormone combining insulin with glucagon or pramlintide) and adjunctive therapy (eg, GLP-1 and SGLT2 inhibitors) also represent promising options. Meal glucose control with the artificial pancreas remains an important challenge for which the optimal strategy is still to be determined.

Model-Free Machine Learning in Biomedicine: Feasibility Study in Type 1 Diabetes

Although reinforcement learning (RL) is suitable for highly uncertain systems, the applicability of this class of algorithms to medical treatment may be limited by the patient variability which dictates individualised tuning for their usually multiple algorithmic parameters. This study explores the feasibility of RL in the framework of artificial pancreas development for type 1 diabetes (T1D). In this approach, an Actor-Critic (AC) learning algorithm is designed and developed for the optimisation of insulin infusion for personalised glucose regulation. AC optimises the daily basal insulin rate and insulin:carbohydrate ratio for each patient, on the basis of his/her measured glucose profile. Automatic, personalised tuning of AC is based on the estimation of information transfer (IT) from insulin to glucose signals. Insulin-to-glucose IT is linked to patient-specific characteristics related to total daily insulin needs and insulin sensitivity (SI). The AC algorithm is evaluated using an FDA-accepted T1D simulator on a large patient database under a complex meal protocol, meal uncertainty and diurnal SI variation. The results showed that 95.66% of time was spent in normoglycaemia in the presence of meal uncertainty and 93.02% when meal uncertainty and SI variation were simultaneously considered. The time spent in hypoglycaemia was 0.27% in both cases. The novel tuning method reduced the risk of severe hypoglycaemia, especially in patients with low SI.