Background and Aims

Interest in artificial pancreas (AP) systems has increased as hybrid-APs have become available for use in daily life. These first-generation systems require manual announcements of meals and exercise, causing management burden to users. Interest in use of additional signals such as heart rate or accelerometer signals to automate and improve AP performance during physical activities is growing. While T1D simulators which accurately model glucose-insulin dynamics exist, they do not provide physiological signals to allow for rapid-prototyping of multivariable AP systems. A multivariable simulator enables rapid advances in multivariable AP research.

Methods

The effect of exercise on glucose dynamics is combined with the Hovorka glucose-insulin dynamic model. Additional models compute physiological signals (heart rate, energy expenditure, and accelerometer values) with their dynamics driven by physical activity. Data for these signals was collected in clinical studies of people with T1D and are used to characterize the interpatient variability and determine realistic parameters for the virtual subjects.

Results

The root mean square error for the original Hovorka model (12.86±6.37mg/dL) is reduced (9.85±5.13mg/dL) when validated against 18 adults with T1D (p=6.4x10^-7). This simulator (mGIPsim) is freely available for academic research. Users of mGIPsim provide the meal and exercise scenarios and insulin infusions. Model equations are then solved to yield the output variables (physiological biosignals, blood glucose concentration, CGM, and plasma insulin concentration).

Conclusions

A multivariable T1D simulator allows for rapid prototyping of physical activity-oriented AP system that can utilize physiological signals such as heart rate or energy expenditure in their algorithms.

Background and Aims

For an artificial pancreas (AP) to effectively limit postprandial glucose excursions, it is necessary that the insulin’s glucose lowering effect starts early in relation to meal onset. Automated and reliable early meal onset detection could therefore enhance the control outcome of APs.

A typical AP depends on continuous glucose monitoring (CGM) for insulin dosing. Because of the slow dynamics of the glucose sensing, current CGM based meal detection approaches typically exhibit a delay of 10 minutes between actual meal onset and reliable detection. In contrast, the processes of ingestion and digestion produce sounds even before meal glucose enters the blood.

Therefore, the focus of the present work is towards the early meal onset detection based on abdominal sounds (AS).

Methods

In this work we employ AS recorded in two healthy volunteers with a condenser microphone, and present an automated approach. We use the Mel-frequency cepstral coefficients and wavelet entropy as features. These features are fed to a feed forward neural network for discriminating the “meal” and “no-meal” classes

Results

This approach detects meal onset with an average delay of 4.3 minutes in our limited number of subjects. Importantly, it provides lesser delay than the state-of-the-art CGM based approach .

Conclusions

Preliminary results indicate that the AS-based approach [1] may provide early meal onset information. This can be exploited in an AP through allowable earlier meal insulin boluses, resulting in improved glycemic control.

References:

Background and Aims

The founder of Inreda Diabetic, manufacturer of a fully-automated bihormonal artificial pancreas system (AP) suffers from type 1 diabetes himself. After further improvements of the AP tested before in ten patients at home, he started AP therapy in November 2018 and has been using it ever since. To our knowledge a bihormonal AP has not yet been used for such a long period.

Methods

This case report describes self-experimentation during daily life. Patient’s treatment before using the AP consisted of multiple daily insulin injections and SMBG. The AP, which uses two sensors that continuously monitor the glucose level and administers insulin or glucagon accordingly, stores the glucose values and sends this to a secured server. HbA1c values were measured during regular hospital visits.

Results

Presented values are median [IQR] values over the period from November 18, 2018 until September 30, 2019. The daily time in range (3.9-10 mmol/l) was 92.6 [88.1 – 96.7]% , the median glucose was 7.1 [6.7 – 7.5] mmol/l. The daily time in hypoglycemia was 0.0 [0.0 – 1.0]%. Last measured HbA1c before AP therapy was 66 mmol/mol. Six weeks after the start of AP therapy the HbA1c was 54 mmol/mol and then remained stable.

Conclusions

The bihormonal AP provides good long-term glucose control in this patient. The improved HbA1c values indicate clear health benefits for this patient. Although long-term data is only available for one patient, we expect that the AP will provide good long-term glucose control in other patients with type 1 diabetes as well.

Background and Aims

Artificial pancreas is considered as the state-of-the-art treatment for type-I diabetes while islets transplantation (IT) is the only curative method. IT is associated with significant drawbacks, primarily tissue availability and mandatory use of immunosuppressive drug therapy. The first could be met by employing stem-cell-derived products or porcine islets, the second – by separating the graft from the host immune system. Using parting approach renders the graft avascular so nutrients and waste products are transferred across the membrane by diffusion only. Exercising minimally invasive surgery, encapsulated islets are implanted into an oxygen-poor subcutaneous site. Islets cells, however, are metabolically active and their functionality is oxygen-dependent.

Methods

To meet regulatory, medical and functional requests, we developed a retrievable BAP macro-device - the βAir. It includes three modules: islets, air chamber, and membrane. Gaseous oxygen is supplied to the islets from the air chamber.

Results

We evaluated the potency of the βAir BAP to cure experimental diabetes in allogeneic small animal models for a period of six months. Xenogeneic islets were implanted into mini-swine and monkeys with remarkable results. Two clinical trials, using minimal islets dose demonstrated clear clinical advantage.

Conclusions

Altogether, the capacity of the technology to achieve close to normal control over blood glucose in diabetic models was demonstrated. The direction towards a commercial BAP is understood. We now present Gen2 BAP in which building materials were optimized; the one-piece device was separated into its components – the islets module and the air chamber. Clinical trials using this device will commence in two years.

Background and Aims

Exercise remains challenging to current generation closed-loop (CL) systems relying on glucose as the sole measured input determining insulin delivery. Interstitial lactate may be an additional signal modulating insulin delivery with exercise. We aimed to explore the feasibility of continuous interstitial lactate measurements and their relationship to venous lactate in individuals with T1D during exercise.

Methods

Six adults with T1D (mean ± SD; age: 40.0 ± 9.5y; HbA_1c: 7.6 ± 1.3%) had a prototype optochemical continuous lactate monitor (CLM) inserted subcutaneously in their flank. All participants undertook 40min high intensity exercise. Forearm venous samples for lactate measurement by YSI analyser were collected at 20min intervals from exercise commencement until 240min post-exercise. Exploratory comparisons between CLM and venous lactate profiles included comparisons of time-to-peak and lag-time.

Results

Preliminary data was analysed from six participants. Unacceptable signal loss occurred in two participants post-exercise, with data only included for time-to-peak analysis. There was high variability in lag-time of CLM versus venous lactate, with a difference in time-to-peak ranging from -11 (faster in CLM) to +47 min (n=6) and lactate clearance ranging from -40 (faster in CLM) to +114min (n=4).

Conclusions

This early data suggests that interstitial lactate as measured by a prototype CLM in T1D participants undertaking high intensity exercise mirrored the rise and fall in venous lactate with variable lag. This lag appeared greater when lactate levels were falling post-exercise. These differences between interstitial and venous lactate may have implications for interstitial lactate as a CL additional signal candidate.

Background and Aims

INTRODUCTION: Diabetes technology improves glycaemic control. However, hypoglycaemia remains the main challenge for optimizing it. Integrated systems such as Do It Yourself (DIY) Artificial Pancreas (AP), known worldwide although not FDA approved, can be a major breakthrough in overcoming this barrier.

Methods

METHODS: We present data from a patient using DIY AP (hybrid closed-loop system) consisting of a Roche Combo pump with faster insulin aspart (Fiasp), linked to Dexcom G6 CGM and an open source app that acts as an AP system (Android APS). Real time access to CGM data is provided via Nightscout Open Source (CGM in the cloud). We compare glycaemic data before and after AP system.

Results

RESULTS: 42-year-old male with type 1 DM. Diabetes duration of 28 years, basal-bolus insulin therapy Degludec/Lispro, with good metabolic control but high frequency of hypoglycaemias. Last 3 months FreeStyle Libre Flash CGM readings showed: Mean of 15 scans/day. Estimated Hba1c 5.5%.228 low glucose events. 104 min of time spent in hypoglycaemia. Time in range (TIR) 77% (70-180 mg/dl), 16% below and 7% above target. Hypoglycaemic events were systematically more than 3/day, several of them <54 mg/dl, various were severe. With AP system, Nightscout 3 months reporting showed a significant improve in TIR (77% to 93.2%), and in low values (16% to 4.0%), these averaged 61.1 +/- 6.9 mg/dl. Estimated Hba1c 5.4%.0 severe hypoglycaemias.

Conclusions

CONCLUSION: Elimination of hypoglycaemia is a major challenge in type 1 DM. AP system shows promising results demonstrating a reduction in hypoglycaemia, improving glycaemic control and quality of life.

Background and Aims

Light-emitting diode (LED) phototherapy regenerates pancreatic islets and alters carbohydrate metabolism Phototherapy has shown good results for cell proliferation and regeneration.We investigated the effects of LED (LED) irradiation in the pancreas injured to induction of experimental diabetes and we evaluated the morphological changes in pancreatic β cells.

Methods

For the experiment, we used twenty randomly selected Wistar rats in three groups: non-diabetic diabetic and diabetic control treated with LED Diabetes was induced by injection of streptozotocin. The irradiated group was treated with LED (λ = 805 nm; 40 mW, 22 s; 0.88 J), applied to the anatomical area of the pancreas for 5 consecutive days and evaluated after 30 days.

Results

Islet and duct regeneration in the pancreas was observed after 30 days in the diabetic group treated with LED, and this regeneration was statistically significant when compared to the control group (p = 0.01). In the diabetic control group, hepatic glycogen content was lower when compared to diabetics with LED (p = 0.03). When performing the intraperitoneal insulin tolerance test, we observed differences between diabetic control and diabetic treatment groups (p = 0.03).

Conclusions

This study demonstrated that LED phototherapy allowed regeneration of pancreatic tissue, especially pancreatic islets and altered carbohydrate metabolism in an experimental model.

Background and Aims

We evaluated the function of the insulin-only iLet bionic pancreas delivering Fiasp vs. iLet delivering insulin lispro or aspart vs. usual care in a home-use study in adults on pump and MDI therapy with type 1 diabetes.

Methods

We performed a 3-way, random-order cross-over, home use study comparing the iLet delivering Fiasp (iLet-F) vs. the iLet delivering insulin lispro or aspart according to the usual-care insulin (iLet-LA) vs. usual care (UC) for 7 days each. Bionic pancreas sessions were initiated by entering only the body weight; the iLet autonomously and continuously adapts to individual insulin needs. The PK setting in the iLet algorithm was not adjusted for Fiasp.

Results

The mean CGMG in the iLet-F arm (155±11, p=0.042), but not in the iLet-LA (155±13, p=0.097), was significantly lower than in the UC arm (162±26). There was no difference in mean CGMG between the iLet-F and iLet-LA arms (p=0.64). There were no differences in median % time <54 mg/dl between the arms (0.49 [0.0,1.0] vs. 0.53 [0.2,1.0] vs. 0.35 [0.1,1.2], p>0.64). The % time in range was greater in the iLet-F (70.6±8.1%, p=0.001) and iLet-LA (70.1±9.2%, p=0.006) arms vs. the UC arm (61.5%). There was no difference between the iLet-F and iLet-LA arms (p=0.54). There were no differences in mean insulin TDD between arms (p>0.45).

Conclusions

The iLet can provide effective glucose control when delivering Fiasp, insulin aspart, or insulin lispro. Adjustment to the PK settings of the iLet may be necessary to further improve glycemic outcomes with Fiasp.

Background and Aims

Intra-individual variability in insulin absorption after subcutaneous (SC) infusion may impair glycemic outcomes and has not been widely studied in the hybrid closed-loop (HCL) setting. Our aim was to ascertain the intra-individual variability in SC insulin absorption by means of a previously validated model (Schiavon et al., IEEE 2018) during a standardized HCL meal study.

Methods

Ten youths with T1D (age=20.9±3.7 y; BMI=23.6±4.5 kg/m²; TDD= 50.6±16.3 U/day) underwent two consecutive, standardized meal studies (70g carbohydrate breakfast and lunch meals) on the same day during DiAs HCL (CGM: Dexcom G4 Platinum with software 505; Insulin pump: Tandem t:slim^TM) treatment. Pre-meal insulin bolus was determined based on subjects’ insulin to carbohydrate ratio and was delivered at the beginning of each meal. Plasma insulin aspart concentrations were measured every 10min for 4h during each meal and were used to estimate, for each subject and for each meal, a set of SC insulin absorption parameters (Figure A). The primary metric to assess intra-individual variability of model-derived parameters was the between-meals coefficient of variation (CV %).

Results

As shown in Figure B, mean values of model parameters have been similar between meals (p=NS); however mean intra-individual variability (CV) ranged from 34-94%, with the highest CV for the model parameter representing the direct absorption of non-monomeric insulin to plasma (k_a1).

Conclusions

Our preliminary results suggest high intra-individual variability in SC insulin absorption during HCL treatment and underline the importance of optimizing insulin delivery algorithms to account for such variability to improve treatment outcomes.

Background and Aims

Advanced closed loop (ACL) algorithms combine automated basal rate with additional enhancement when correction is required, usually post meal. Preliminary studies have demonstrated increased time spent in target glucose range of 70-180 mg/dL (TIR) with reduction in post prandial excursion. Here, we assessi the effectiveness of the algorithm to overcome no premeal bolus.

Methods

Four participants were followed for 4 days in a protected free living environment, while consuming pre-defined meals consisting of either 40, 60 and 80 grams of carbohydrates each day. Participants consumed the same meals and either bolus or did not bolus before consuming the meals according to the following protocol: Day1, all meals announced. Day 2 , participants announce only the 80 gr carb meal. In days 3 and 4, all meals unannounced

Results

Preliminary results of 4 adult participants show that overall glycemia during the unannounced meal phase demonstrated a 73.2% (±9.6) TIR of 70-180 mg% and 0% time in hypoglycemia <70mg%. Reduction in total daily dose and requirements for glucose salvage was noted. Comparison between the post meal excursion of the announced versus unannounced meals demonstrated a separation of excursion, when the carbs was above 40 grams. The peak glucose levels of unannounced meals did not differ between the 40, 60 and 80 gram carbohydrate-containing meals,

Conclusions

MINIMED^TM 670G 4.0 system is programmed for meal announcement. Nevertheless, when meals containing < 80 gram of carbohydrates are consumed without meal announcement, the system is able to provide safe glycemic control with over 70% of TIR.

Background and Aims

An Artificial Pancreas usually involves subcutaneous (sc) devices to measure glucose and to infuse insulin with important delays. The intraperitoneal (ip) route is more physiological and would avoid the sc delays, improving glucose control[A]. In these last years, new ip devices have been designed[B] and new modalities of ip insulin delivery, e.g pulsatile, have been tested. The aim here is to present the new simulator and its first use in in silico trial comparing the ip vs. sc insulin delivery route.

[A]Dassau, E. et al. Diabetes,Obesity and Metabolism,19.12(2017):1698-1705.

[B]Iacovacci V. et al. Journal of Medical Devices,13.1(2019):011008.

Methods

The model of the UVA/Padova simulator has been modified to describe ip insulin administration. A Run-to-Run approach is used to adapt the Basal-Bolus Therapy (BBT) of the 100 in silico patients to the new ip site. A new version of the MPC controller [C] is synthesised using the new simulator and the optimized BBT.

[C] Soru P. et al. Annual Reviews in Control,36.1(2012):118-128.

Results

The intraperitoneal controller tested on a 2 days scenario maintains the glucose inside the target range for 93.5% of the time with no time spent below 70 mg/dl and 6.5% above 180 mg/dl. The MPC is able to keep the glycaemia inside the target range during all night.

Conclusions

The ip MPC results are much better than those obtained with the sc MPC. Future development of MPC involves the design of ad hoc constraints and safety. This new MPC in conjunctions with a new ip insulin pump will be tested in animals.

Background and Aims

Maintaining glucose concentration (GC) in the target range in spite of physical activities and events causing acute psychological stress (PS) is challenging for artificial pancreas (AP) systems. PS and physical activities affect GC in different ways; aerobic exercise decreases GC while PS can increase it. Many stressful events and physical activities cannot be manually entered to the AP in a timely manner. Hence, they are unknown disturbances for AP systems and reduce glycemic control. In this work, wristband biosignals are utilized in a novel algorithm to estimate the psychological and physiological state of a subject, and improve the GC prediction for use in AP systems.

Methods

Biosignals from Empatica E4 wristband are collected in real-time and machine learning algorithms are utilized to determine the physical state of a subject, obtain energy expenditure estimates, and predict her/his PS levels. These estimates are incorporated in a GC prediction model along with CGM readings and insulin infusion data from pump. These BGC estimates are compared to estimates from a model that uses only CGM readings and insulin infusion data.

Results

Data from 50 experiments with thirteen different subjects with T1D who performed physical activities and various PS causing events were used and proposed method improved to mean absolute percentage error of GC prediction by 6.5%.

Conclusions

Wristband biosignals used to determine the psychological and physiological state of people with T1D provide valuable information to improve GC estimates and the performance of an AP system in response to unannounced physical activities and stressful events.