Associations between daily step count classifications and continuous glucose monitoring metrics in adults with type 1 diabetes: analysis of the Type 1 Diabetes Exercise Initiative (T1DEXI) cohort

AIMS/HYPOTHESIS

Adults with type 1 diabetes should perform daily physical activity to help maintain health and fitness, but the influence of daily step counts on continuous glucose monitoring (CGM) metrics are unclear. This analysis used the Type 1 Diabetes Exercise Initiative (T1DEXI) dataset to investigate the effect of daily step count on CGM-based metrics.

METHODS

In a 4 week free-living observational study of adults with type 1 diabetes, with available CGM and step count data, we categorised participants into three groups-below (<7000), meeting (7000-10,000) or exceeding (>10,000) the daily step count goal-to determine if step count category influenced CGM metrics, including per cent time in range (TIR: 3.9-10.0 mmol/l), time below range (TBR: <3.9 mmol/l) and time above range (TAR: >10.0 mmol/l).

RESULTS

A total of 464 adults with type 1 diabetes (mean±SD age 37±14 years; HbA1c 48.8±8.1 mmol/mol [6.6±0.7%]; 73% female; 45% hybrid closed-loop system, 38% standard insulin pump, 17% multiple daily insulin injections) were included in the study. Between-participant analyses showed that individuals who exceeded the mean daily step count goal over the 4 week period had a similar TIR (75±14%) to those meeting (74±14%) or below (75±16%) the step count goal (p>0.05). In the within-participant comparisons, TIR was higher on days when the step count goal was exceeded or met (both 75±15%) than on days below the step count goal (73±16%; both p<0.001). The TBR was also higher when individuals exceeded the step count goals (3.1%±3.2%) than on days when they met or were below step count goals (difference in means -0.3% [p=0.006] and -0.4% [p=0.001], respectively). The total daily insulin dose was lower on days when step count goals were exceeded (0.52±0.18 U/kg; p<0.001) or were met (0.53±0.18 U/kg; p<0.001) than on days when step counts were below the current recommendation (0.55±0.18 U/kg). Step count had a larger effect on CGM-based metrics in participants with a baseline HbA1c ≥53 mmol/mol (≥7.0%).

CONCLUSIONS/INTERPRETATION

Our results suggest that, compared with days with low step counts, days with higher step counts are associated with slight increases in both TIR and TBR, along with small reductions in total daily insulin requirements, in adults living with type 1 diabetes.

DATA AVAILABILITY

The data that support the findings reported here are available on the Vivli Platform (ID: T1-DEXI; https://doi.org/10.25934/PR00008428 ).

Digital twins and artificial intelligence in metabolic disease research

Digital twin technology is emerging as a transformative paradigm for personalized medicine in the management of chronic conditions. In this article, we explore the concept and key characteristics of a digital twin and its applications in chronic non-communicable metabolic disease management, with a focus on diabetes case studies. We cover various types of digital twin models, including mechanistic models based on ODEs, data-driven ML algorithms, and hybrid modeling strategies that combine the strengths of both approaches. We present successful case studies demonstrating the potential of digital twins in improving glucose outcomes for individuals with T1D and T2D, and discuss the benefits and challenges of translating digital twin research applications to clinical practice.

The Association Between Diet Quality and Glycemic Outcomes Among People with Type 1 Diabetes

Background

The amount and type of food consumed impacts the glycemic response and insulin needs of people with type 1 diabetes mellitus (T1DM). Daily variability in consumption, reflected in diet quality, may acutely impact glycemic levels and insulin needs.

Objective

Type 1 Diabetes Exercise Initiative (T1DEXI) data were examined to evaluate the impact of daily diet quality on near-term glycemic control and interaction with exercise.

Methods

Using the Remote Food Photography Method, ≤8 d of dietary intake data were analyzed per participant. Diet quality was quantified with the Healthy Eating Index-2015 (HEI), where a score of 100 indicates the highest-quality diet. Each participant day was classified as low HEI (≤57) or high HEI (>57) based on the mean of nationally reported HEI data. Within participants, the relationship between diet quality and subsequent glycemia measured by continuous glucose monitoring (CGM) and total insulin dose usage was evaluated using a paired t-test and robust regression models.

Results

Two hundred twenty-three adults (76% female) with mean ± SD age, HbA1c, and body mass index (BMI) of 37 ± 14 y, 6.6% ± 0.7%, and 25.1 ± 3.6 kg/m2, respectively, were included in these analyses. The mean HEI score was 56 across all participant days. On high HEI days (mean, 66 ± 4) compared with low HEI days (mean, 47 ± 5), total time in range (70-180 mg/dL) was greater (77.2% ± 14% compared with 75.7% ± 14%, respectively, P = 0.01), whereas time above 180 mg/dL (19% ± 14% compared with 21% ± 15%, respectively, P = 0.004), mean glucose (143 ± 22 compared with 145 ± 22 mg/dL, respectively, P = 0.02), and total daily insulin dose (0.52 ± 0.18 compared with 0.54 ± 0.18 U/kg/d, respectively, P = 0.009) were lower. The interaction between diet quality and exercise on glycemia was not significant.

Conclusions

Higher HEI scores correlated with improved glycemia and lower insulin needs, although the impact of diet quality was modest and smaller than the previously reported impact of exercise.

Factors Affecting Reproducibility of Change in Glucose During Exercise: Results From the Type 1 Diabetes and EXercise Initiative

AIMS

To evaluate factors affecting within-participant reproducibility in glycemic response to different forms of exercise.

METHODS

Structured exercise sessions ~30 minutes in length from the Type 1 Diabetes Exercise Initiative (T1DEXI) study were used to assess within-participant glycemic variability during and after exercise. The effect of several pre-exercise factors on the within-participant glycemic variability was evaluated.

RESULTS

Data from 476 adults with type 1 diabetes were analyzed. A participant's change in glucose during exercise was reproducible within 15 mg/dL of the participant's other exercise sessions only 32% of the time. Participants who exercised with lower and more consistent glucose level, insulin on board (IOB), and carbohydrate intake at exercise start had less variability in glycemic change during exercise. Participants with lower mean glucose (P < .001), lower glucose coefficient of variation (CV) (P < .001), and lower % time <70 mg/dL (P = .005) on sedentary days had less variable 24-hour post-exercise mean glucose.

CONCLUSIONS

Reproducibility of change in glucose during exercise was low in this cohort of adults with T1D, but more consistency in pre-exercise glucose levels, IOB, and carbohydrates may increase this reproducibility. Mean glucose variability in the 24 hours after exercise is influenced more by the participant's overall glycemic control than other modifiable factors.

Design and In Silico Evaluation of an Exercise Decision Support System Using Digital Twin Models

BACKGROUND

Managing glucose levels during exercise is challenging for individuals with type 1 diabetes (T1D) since multiple factors including activity type, duration, intensity and other factors must be considered. Current decision support tools lack personalized recommendations and fail to distinguish between aerobic and resistance exercise. We propose an exercise-aware decision support system (exDSS) that uses digital twins to deliver personalized recommendations to help people with T1D maintain safe glucose levels (70-180 mg/dL) and avoid low glucose (<70 mg/dL) during and after exercise.

METHODS

We evaluated exDSS using various exercise and meal scenarios recorded from a large, free-living study of aerobic and resistance exercise. The model inputs were heart rate, insulin, and meal data. Glucose responses were simulated during and after 30-minute exercise sessions (676 aerobic, 631 resistance) from 247 participants. Glucose outcomes were compared when participants followed exDSS recommendations, clinical guidelines, or did not modify behavior (no intervention).

RESULTS

exDSS significantly improved mean time in range for aerobic (80.2% to 92.3%, P < .0001) and resistance (72.3% to 87.3%, P < .0001) exercises compared with no intervention, and versus clinical guidelines (aerobic: 82.2%, P < .0001; resistance: 80.3%, P < .0001). exDSS reduced time spent in low glucose for both exercise types compared with no intervention (aerobic: 15.1% to 5.1%, P < .0001; resistance: 18.2% to 6.6%, P < .0001) and was comparable with following clinical guidelines (aerobic: 4.5%, resistance: 8.1%, P = N.S.).

CONCLUSIONS

The exDSS tool significantly improved glucose outcomes during and after exercise versus following clinical guidelines and no intervention providing motivation for clinical evaluation of the exDSS system.

Artificial Intelligence and Machine Learning for Improving Glycemic Control in Diabetes: Best Practices, Pitfalls, and Opportunities

OBJECTIVE

Artificial intelligence and machine learning are transforming many fields including medicine. In diabetes, robust biosensing technologies and automated insulin delivery therapies have created a substantial opportunity to improve health. While the number of manuscripts addressing the topic of applying machine learning to diabetes has grown in recent years, there has been a lack of consistency in the methods, metrics, and data used to train and evaluate these algorithms. This manuscript provides consensus guidelines for machine learning practitioners in the field of diabetes, including best practice recommended approaches and warnings about pitfalls to avoid.

METHODS

Algorithmic approaches are reviewed and benefits of different algorithms are discussed including importance of clinical accuracy, explainability, interpretability, and personalization. We review the most common features used in machine learning applications in diabetes glucose control and provide an open-source library of functions for calculating features, as well as a framework for specifying data sets using data sheets. A review of current data sets available for training algorithms is provided as well as an online repository of data sources.

SIGNIFICANCE

These consensus guidelines are designed to improve performance and translatability of new machine learning algorithms developed in the field of diabetes for engineers and data scientists.

Combining uncertainty-aware predictive modeling and a bedtime Smart Snack intervention to prevent nocturnal hypoglycemia in people with type 1 diabetes on multiple daily injections

OBJECTIVE

Nocturnal hypoglycemia is a known challenge for people with type 1 diabetes, especially for physically active individuals or those on multiple daily injections. We developed an evidential neural network (ENN) to predict at bedtime the probability and timing of nocturnal hypoglycemia (0-4 vs 4-8 h after bedtime) based on several glucose metrics and physical activity patterns. We utilized these predictions in silico to prescribe bedtime carbohydrates with a Smart Snack intervention specific to the predicted minimum nocturnal glucose and timing of nocturnal hypoglycemia.

MATERIALS AND METHODS

We leveraged free-living datasets collected from 366 individuals from the T1DEXI Study and Glooko. Inputs to the ENN used to model nocturnal hypoglycemia were derived from demographic information, continuous glucose monitoring, and physical activity data. We assessed the accuracy of the ENN using area under the receiver operating curve, and the clinical impact of the Smart Snack intervention through simulations.

RESULTS

The ENN achieved an area under the receiver operating curve of 0.80 and 0.71 to predict nocturnal hypoglycemic events during 0-4 and 4-8 h after bedtime, respectively, outperforming all evaluated baseline methods. Use of the Smart Snack intervention reduced probability of nocturnal hypoglycemia from 23.9 ± 14.1% to 14.0 ± 13.3% and duration from 7.4 ± 7.0% to 2.4 ± 3.3% in silico.

DISCUSSION

Our findings indicate that the ENN-based Smart Snack intervention has the potential to significantly reduce the frequency and duration of nocturnal hypoglycemic events.

CONCLUSION

A decision support system that combines prediction of minimum nocturnal glucose and proactive recommendations for bedtime carbohydrate intake might effectively prevent nocturnal hypoglycemia and reduce the burden of glycemic self-management.

Risky movement: Assessing fall risk in people with multiple sclerosis with wearable sensors and beacon-based smart-home monitoring

BACKGROUND

People with multiple sclerosis (PwMS) fall frequently causing injury, social isolation, and decreased quality of life. Identifying locations and behaviors associated with high fall risk could help direct fall prevention interventions. Here we describe a smart-home system for assessing how mobility metrics relate to real-world fall risk in PwMS.

METHODS

We performed a secondary analysis of a dataset of real-world falls collected from PwMS to identify patterns associated with increased fall risk. Thirty-four individuals were tracked over eight weeks with an inertial sensor comprising a triaxial accelerometer and time-of-flight radio transmitter, which communicated with beacons positioned throughout the home. We evaluated associations between locations in the home and movement behaviors prior to a fall compared with time periods when no falls occurred using metrics including gait initiation, time-spent-moving, movement length, and an entropy-based metric that quantifies movement complexity using transitions between rooms in the home. We also explored how fall risk may be related to the percent of times that a participant paused while walking (pauses-while-walking).

RESULTS

Seventeen of the participants monitored sustained a total of 105 falls that were recorded. More falls occurred while walking (52%) than when stationary despite participants being largely sedentary, only walking 1.5±3.3% (median ± IQR) of the time that they were in their home. A total of 28% of falls occurred within one second of gait initiation. As the percentage of pauses-while-walking increased from 20 to 60%, the likelihood of a fall increased by nearly 3 times from 0.06 to 0.16%. Movement complexity, which was quantified using the entropy of room transitions, was significantly higher in the 10 min preceding falls compared with other 10-min time segments not preceding falls (1.15 ± 0.47 vs. 0.96 ± 0.24, P = 0.02). Path length was significantly longer (151.3 ± 156.1 m vs. 95.0 ± 157.2 m, P = 0.003) in the ten minutes preceding a fall compared with non-fall periods. Fall risk also varied among rooms but not consistently across participants.

CONCLUSIONS

Movement metrics derived from wearable sensors and smart-home tracking systems are associated with fall risk in PwMS. More pauses-while-walking, and more complex, longer movement trajectories are associated with increased fall risk.

FUNDING

Department of Veterans Affairs (RX001831-01A1). National Science Foundation (#2030859).

Quantifying insulin-mediated and noninsulin-mediated changes in glucose dynamics during resistance exercise in type 1 diabetes

Exercise can cause dangerous fluctuations in blood glucose in people living with type 1 diabetes (T1D). Aerobic exercise, for example, can cause acute hypoglycemia secondary to increased insulin-mediated and noninsulin-mediated glucose utilization. Less is known about how resistance exercise (RE) impacts glucose dynamics. Twenty-five people with T1D underwent three sessions of either moderate or high-intensity RE at three insulin infusion rates during a glucose tracer clamp. We calculated time-varying rates of endogenous glucose production (EGP) and glucose disposal (Rd) across all sessions and used linear regression and extrapolation to estimate insulin- and noninsulin-mediated components of glucose utilization. Blood glucose did not change on average during exercise. The area under the curve (AUC) for EGP increased by 1.04 mM during RE (95% CI: 0.65-1.43, P < 0.001) and decreased proportionally to insulin infusion rate (0.003 mM per percent above basal rate, 95% CI: 0.001-0.006, P = 0.003). The AUC for Rd rose by 1.26 mM during RE (95% CI: 0.41-2.10, P = 0.004) and increased proportionally with insulin infusion rate (0.04 mM per percent above basal rate, CI: 0.03-0.04, P < 0.001). No differences were observed between the moderate and high resistance groups. Noninsulin-mediated glucose utilization rose significantly during exercise before returning to baseline roughly 30-min postexercise. Insulin-mediated glucose utilization remained unchanged during exercise sessions. Circulating catecholamines and lactate rose during exercise despite relatively small changes observed in Rd. Results provide an explanation of why RE may pose a lower overall risk for hypoglycemia.NEW & NOTEWORTHY Aerobic exercise is known to cause decreases in blood glucose secondary to increased glucose utilization in people living with type 1 diabetes (T1D). However, less is known about how resistance-type exercise impacts glucose dynamics. Twenty-five participants with T1D performed in-clinic weight-bearing exercises under a glucose clamp. Mathematical modeling of infused glucose tracer allowed for quantification of the rate of hepatic glucose production as well as rates of insulin-mediated and noninsulin-mediated glucose uptake experienced during resistance exercise.

The Type 1 Diabetes and EXercise Initiative: Predicting Hypoglycemia Risk During Exercise for Participants with Type 1 Diabetes Using Repeated Measures Random Forest

Objective: Exercise is known to increase the risk for hypoglycemia in type 1 diabetes (T1D) but predicting when it may occur remains a major challenge. The objective of this study was to develop a hypoglycemia prediction model based on a large real-world study of exercise in T1D. Research Design and Methods: Structured study-specified exercise (aerobic, interval, and resistance training videos) and free-living exercise sessions from the T1D Exercise Initiative study were used to build a model for predicting hypoglycemia, a continuous glucose monitoring value <70 mg/dL, during exercise. Repeated measures random forest (RMRF) and repeated measures logistic regression (RMLR) models were constructed to predict hypoglycemia using predictors at the start of exercise and baseline characteristics. Models were evaluated with area under the receiver operating characteristic curve (AUC) and balanced accuracy. Results: RMRF and RMLR had similar AUC (0.833 vs. 0.825, respectively) and both models had a balanced accuracy of 77%. The probability of hypoglycemia was higher for exercise sessions with lower pre-exercise glucose levels, negative pre-exercise glucose rates of change, greater percent time <70 mg/dL in the 24 h before exercise, and greater pre-exercise bolus insulin-on-board (IOB). Free-living aerobic exercises, walking/hiking, and physical labor had the highest probability of hypoglycemia, while structured exercises had the lowest probability of hypoglycemia. Conclusions: RMRF and RMLR accurately predict hypoglycemia during exercise and identify factors that increase the risk of hypoglycemia. Lower glucose, decreasing levels of glucose before exercise, and greater pre-exercise IOB largely predict hypoglycemia risk in adults with T1D.

Integrating metabolic expenditure information from wearable fitness sensors into an AI-augmented automated insulin delivery system: a randomised clinical trial

BACKGROUND

Exercise can rapidly drop glucose in people with type 1 diabetes. Ubiquitous wearable fitness sensors are not integrated into automated insulin delivery (AID) systems. We hypothesised that an AID can automate insulin adjustments using real-time wearable fitness data to reduce hypoglycaemia during exercise and free-living conditions compared with an AID not automating use of fitness data.

METHODS

Our study population comprised of individuals (aged 21-50 years) with type 1 diabetes from from the Harold Schnitzer Diabetes Health Center clinic at Oregon Health and Science University, OR, USA, who were enrolled into a 76 h single-centre, two-arm randomised (4-block randomisation), non-blinded crossover study to use (1) an AID that detects exercise, prompts the user, and shuts off insulin during exercise using an exercise-aware adaptive proportional derivative (exAPD) algorithm or (2) an AID that automates insulin adjustments using fitness data in real-time through an exercise-aware model predictive control (exMPC) algorithm. Both algorithms ran on iPancreas comprising commercial glucose sensors, insulin pumps, and smartwatches. Participants executed 1 week run-in on usual therapy followed by exAPD or exMPC for one 12 h primary in-clinic session involving meals, exercise, and activities of daily living, and 2 free-living out-patient days. Primary outcome was time below range (<3·9 mmol/L) during the primary in-clinic session. Secondary outcome measures included mean glucose and time in range (3·9-10 mmol/L). This trial is registered with ClinicalTrials.gov, NCT04771403.

FINDINGS

Between April 13, 2021, and Oct 3, 2022, 27 participants (18 females) were enrolled into the study. There was no significant difference between exMPC (n=24) versus exAPD (n=22) in time below range (mean [SD] 1·3% [2·9] vs 2·5% [7·0]) or time in range (63·2% [23·9] vs 59·4% [23·1]) during the primary in-clinic session. In the 2 h period after start of in-clinic exercise, exMPC had significantly lower mean glucose (7·3 [1·6] vs 8·0 [1·7] mmol/L, p=0·023) and comparable time below range (1·4% [4·2] vs 4·9% [14·4]). Across the 76 h study, both algorithms achieved clinical time in range targets (71·2% [16] and 75·5% [11]) and time below range (1·0% [1·2] and 1·3% [2·2]), significantly lower than run-in period (2·4% [2·4], p=0·0004 vs exMPC; p=0·012 vs exAPD). No adverse events occurred.

INTERPRETATION

AIDs can integrate exercise data from smartwatches to inform insulin dosing and limit hypoglycaemia while improving glucose outcomes. Future AID systems that integrate exercise metrics from wearable fitness sensors may help people living with type 1 diabetes exercise safely by limiting hypoglycaemia.

FUNDING

JDRF Foundation and the Leona M and Harry B Helmsley Charitable Trust, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases.

Predicting labor onset relative to the estimated date of delivery using smart ring physiological data

The transition from pregnancy into parturition is physiologically directed by maternal, fetal and placental tissues. We hypothesize that these processes may be reflected in maternal physiological metrics. We enrolled pregnant participants in the third-trimester (n = 118) to study continuously worn smart ring devices monitoring heart rate, heart rate variability, skin temperature, sleep and physical activity from negative temperature coefficient, 3-D accelerometer and infrared photoplethysmography sensors. Weekly surveys assessed labor symptoms, pain, fatigue and mood. We estimated the association between each metric, gestational age, and the likelihood of a participant's labor beginning prior to (versus after) the clinical estimated delivery date (EDD) of 40.0 weeks with mixed effects regression. A boosted random forest was trained on the physiological metrics to predict pregnancies that naturally passed the EDD versus undergoing onset of labor prior to the EDD. Here we report that many raw sleep, activity, pain, fatigue and labor symptom metrics are correlated with gestational age. As gestational age advances, pregnant individuals have lower resting heart rate 0.357 beats/minute/week, 0.84 higher heart rate variability (milliseconds) and shorter durations of physical activity and sleep. Further, random forest predictions determine pregnancies that would pass the EDD with accuracy of 0.71 (area under the receiver operating curve). Self-reported symptoms of labor correlate with increased gestational age and not with the timing of labor (relative to EDD) or onset of spontaneous labor. The use of maternal smart ring-derived physiological data in the third-trimester may improve prediction of the natural duration of pregnancy relative to the EDD.

Descriptive Characterization of High-Frequency Distortion Product Otoacoustic Emission Source Components in Children

PURPOSE

Distortion product otoacoustic emissions (DPOAEs) provide an objective assessment of cochlear function and are used for serial ototoxicity monitoring in pediatric cancer patients. DPOAEs are modeled as having distortion (near f2) and reflection (near 2f1-f2) component sources, and developmental changes are observed in these components' relative strengths in infants compared with adults. However, little is known about source component strengths in childhood or at extended high frequencies (EHFs; > 8 kHz). Thus, the purpose of this study was to describe the effects of age and stimulus frequency on DPOAE components in children.

METHOD

DPOAEs were collected with varied frequency ratios (f2/f1 = 1.1-1.25) for a wide range of frequencies (2-16 kHz) in 39 younger (3-6 years) and 41 older (10-12 years) children with constant levels (L1/L2) of 65/50 dB SPL. A depth-compensated simulator sound pressure level method of calibration was employed. A time waveform representation of the results across various ratios was created to estimate peak pressures and latencies of each DPOAE component.

RESULTS

Estimated peak pressures of DPOAE components revealed the greatest differences in DPOAE sources between children occurring at the highest frequencies tested, where the peak pressure of both components was largest for younger compared with older children. Latency differences between the children were only noted at higher frequencies for the distortion component.

CONCLUSIONS

These results suggest that DPOAE levels decrease with age and reflection emissions are vulnerable to cochlear change. This work guides optimization of protocols for pediatric ototoxicity monitoring, whereby including EHF otoacoustic emissions is clearly warranted and choosing to isolate DPOAE sources may prove beneficial.

SUPPLEMENTAL MATERIAL

https://doi.org/10.23641/asha.23669214.

Diabetes Technology Meeting 2022

Diabetes Technology Society hosted its annual Diabetes Technology Meeting from November 3 to November 5, 2022. Meeting topics included (1) the measurement of glucose, insulin, and ketones; (2) virtual diabetes care; (3) metrics for managing diabetes and predicting outcomes; (4) integration of continuous glucose monitor data into the electronic health record; (5) regulation of diabetes technology; (6) digital health to nudge behavior; (7) estimating carbohydrates; (8) fully automated insulin delivery systems; (9) hypoglycemia; (10) novel insulins; (11) insulin delivery; (12) on-body sensors; (13) continuous glucose monitoring; (14) diabetic foot ulcers; (15) the environmental impact of diabetes technology; and (16) spinal cord stimulation for painful diabetic neuropathy. A live demonstration of a device that can allow for the recycling of used insulin pens was also presented.

Examining the Acute Glycemic Effects of Different Types of Structured Exercise Sessions in Type 1 Diabetes in a Real-World Setting: The Type 1 Diabetes and Exercise Initiative (T1DEXI)

OBJECTIVE

Maintenance of glycemic control during and after exercise remains a major challenge for individuals with type 1 diabetes. Glycemic responses to exercise may differ by exercise type (aerobic, interval, or resistance), and the effect of activity type on glycemic control after exercise remains unclear.

RESEARCH DESIGN AND METHODS

The Type 1 Diabetes Exercise Initiative (T1DEXI) was a real-world study of at-home exercise. Adult participants were randomly assigned to complete six structured aerobic, interval, or resistance exercise sessions over 4 weeks. Participants self-reported study and nonstudy exercise, food intake, and insulin dosing (multiple daily injection [MDI] users) using a custom smart phone application and provided pump (pump users), heart rate, and continuous glucose monitoring data.

RESULTS

A total of 497 adults with type 1 diabetes (mean age ± SD 37 ± 14 years; mean HbA1c ± SD 6.6 ± 0.8% [49 ± 8.7 mmol/mol]) assigned to structured aerobic (n = 162), interval (n = 165), or resistance (n = 170) exercise were analyzed. The mean (± SD) change in glucose during assigned exercise was -18 ± 39, -14 ± 32, and -9 ± 36 mg/dL for aerobic, interval, and resistance, respectively (P < 0.001), with similar results for closed-loop, standard pump, and MDI users. Time in range 70-180 mg/dL (3.9-10.0 mmol/L) was higher during the 24 h after study exercise when compared with days without exercise (mean ± SD 76 ± 20% vs. 70 ± 23%; P < 0.001).

CONCLUSIONS

Adults with type 1 diabetes experienced the largest drop in glucose level with aerobic exercise, followed by interval and resistance exercise, regardless of insulin delivery modality. Even in adults with well-controlled type 1 diabetes, days with structured exercise sessions contributed to clinically meaningful improvement in glucose time in range but may have slightly increased time below range.

Enabling fully automated insulin delivery through meal detection and size estimation using Artificial Intelligence

We present a robust insulin delivery system that includes automated meal detection and carbohydrate content estimation using machine learning for meal insulin dosing called robust artificial pancreas (RAP). We conducted a randomized, single-center crossover trial to compare postprandial glucose control in the four hours following unannounced meals using a hybrid model predictive control (MPC) algorithm and the RAP system. The RAP system includes a neural network model to automatically detect meals and deliver a recommended meal insulin dose. The meal detection algorithm has a sensitivity of 83.3%, false discovery rate of 16.6%, and mean detection time of 25.9 minutes. While there is no significant difference in incremental area under the curve of glucose, RAP significantly reduces time above range (glucose >180 mg/dL) by 10.8% (P = 0.04) and trends toward increasing time in range (70-180 mg/dL) by 9.1% compared with MPC. Time below range (glucose <70 mg/dL) is not significantly different between RAP and MPC.

Modeling risk of hypoglycemia during and following physical activity in people with type 1 diabetes using explainable mixed-effects machine learning

BACKGROUND

Physical activity (PA) can cause increased hypoglycemia (glucose <70 mg/dL) risk in people with type 1 diabetes (T1D). We modeled the probability of hypoglycemia during and up to 24 h following PA and identified key factors associated with hypoglycemia risk.

METHODS

We leveraged a free-living dataset from Tidepool comprised of glucose measurements, insulin doses, and PA data from 50 individuals with T1D (6448 sessions) for training and validating machine learning models. We also used data from the T1Dexi pilot study that contains glucose management and PA data from 20 individuals with T1D (139 session) for assessing the accuracy of the best performing model on an independent test dataset. We used mixed-effects logistic regression (MELR) and mixed-effects random forest (MERF) to model hypoglycemia risk around PA. We identified risk factors associated with hypoglycemia using odds ratio and partial dependence analysis for the MELR and MERF models, respectively. Prediction accuracy was measured using the area under the receiver operating characteristic curve (AUROC).

RESULTS

The analysis identified risk factors significantly associated with hypoglycemia during and following PA in both MELR and MERF models including glucose and body exposure to insulin at the start of PA, low blood glucose index 24 h prior to PA, and PA intensity and timing. Both models showed overall hypoglycemia risk peaking 1 h after PA and again 5-10 h after PA, which is consistent with the hypoglycemia risk pattern observed in the training dataset. Time following PA impacted hypoglycemia risk differently across different PA types. Accuracy of hypoglycemia prediction using the fixed effects of the MERF model was highest when predicting hypoglycemia during the first hour following the start of PA (AUROCVALIDATION = 0.83 and AUROCTESTING = 0.86) and decreased when predicting hypoglycemia in the 24 h after PA (AUROCVALIDATION = 0.66 and AUROCTESTING = 0.68).

CONCLUSION

Hypoglycemia risk after the start of PA can be modeled using mixed-effects machine learning to identify key risk factors that may be used within decision support and insulin delivery systems. We published the population-level MERF model online for others to use.

Assessment of a Decision Support System for Adults with Type 1 Diabetes on Multiple Daily Insulin Injections

Introduction: DailyDose is a decision support system designed to provide real-time dosing advice and weekly insulin dose adjustments for adults living with type 1 diabetes using multiple daily insulin injections. Materials and Methods: Twenty-five adults were enrolled in this single-arm study. All participants used Dexcom G6 for continuous glucose monitoring, InPen for short-acting insulin doses, and Clipsulin to track long-acting insulin doses. Participants used DailyDose on an iPhone for 8 weeks. The primary endpoint was % time in range (TIR) comparing the 2-week baseline to the final 2-week period of DailyDose use. Results: There were no significant differences between TIR or other glycemic metrics between the baseline period compared to final 2-week period of DailyDose use. TIR significantly improved by 6.3% when more than half of recommendations were accepted and followed compared with 50% or fewer recommendations (95% CI 2.5%-10.1%, P = 0.001). Conclusions: Use of DailyDose did not improve glycemic outcomes compared to the baseline period. In a post hoc analysis, accepting and following recommendations from DailyDose was associated with improved TIR. Clinical Trial Registration Number: NCT04428645.

Quantifying the impact of physical activity on future glucose trends using machine learning

Prevention of hypoglycemia (glucose <70 mg/dL) during aerobic exercise is a major challenge in type 1 diabetes. Providing predictions of glycemic changes during and following exercise can help people with type 1 diabetes avoid hypoglycemia. A unique dataset representing 320 days and 50,000 + time points of glycemic measurements was collected in adults with type 1 diabetes who participated in a 4-arm crossover study evaluating insulin-pump therapies, whereby each participant performed eight identically designed in-clinic exercise studies. We demonstrate that even under highly controlled conditions, there is considerable intra-participant and inter-participant variability in glucose outcomes during and following exercise. Participants with higher aerobic fitness exhibited significantly lower minimum glucose and steeper glucose declines during exercise. Adaptive, personalized machine learning (ML) algorithms were designed to predict exercise-related glucose changes. These algorithms achieved high accuracy in predicting the minimum glucose and hypoglycemia during and following exercise sessions, for all fitness levels.

Incorporating Glucose Variability into Glucose Forecasting Accuracy Assessment Using the New Glucose Variability Impact Index and the Prediction Consistency Index: An LSTM Case Example

BACKGROUND

In this work, we developed glucose forecasting algorithms trained and evaluated on a large dataset of free-living people with type 1 diabetes (T1D) using closed-loop (CL) and sensor-augmented pump (SAP) therapies; and we demonstrate how glucose variability impacts accuracy. We introduce the glucose variability impact index (GVII) and the glucose prediction consistency index (GPCI) to assess the accuracy of prediction algorithms.

METHODS

A long-short-term-memory (LSTM) neural network was designed to predict glucose up to 60 minutes in the future using continuous glucose measurements and insulin data collected from 175 people with T1D (41,318 days) and evaluated on 75 people (11,333 days) from the Tidepool Big Data Donation Dataset. LSTM was compared with two naïve forecasting algorithms as well as Ridge linear regression and a random forest using root-mean-square error (RMSE). Parkes error grid quantified clinical accuracy. Regression analysis was used to derive the GVII and GPCI.

RESULTS

The LSTM had highest accuracy and best GVII and GPCI. RMSE for CL was 19.8 ± 3.2 and 33.2 ± 5.4 mg/dL for 30- and 60-minute prediction horizons, respectively. RMSE for SAP was 19.6 ± 3.8 and 33.1 ± 7.3 mg/dL for 30- and 60-minute prediction horizons, respectively; 99.6% and 97.6% of predictions were within zones A+B of the Parkes error grid at 30- and 60-minute prediction horizons, respectively. Glucose variability was strongly correlated with RMSE (R≥0.64, P < 0.001); GVII and GPCI demonstrated a means to compare algorithms across datasets with different glucose variability.

CONCLUSIONS

The LSTM model was accurate on a large real-world free-living dataset. Glucose variability should be considered when assessing prediction accuracy using indices such as GVII and GPCI.

Free From Falls education and exercise program for reducing falls in people with multiple sclerosis: A randomized controlled trial

BACKGROUND

People with multiple sclerosis (PwMS) fall frequently. Community-delivered exercise and education reduce falls in older adults, but their efficacy in multiple sclerosis (MS) is unknown.

OBJECTIVES

To evaluate the impact of the Free From Falls (FFF) group education and exercise program on falls in PwMS.

METHODS

This was a prospective, assessor-blinded, two-arm parallel randomized controlled trial. Ninety-six participants were randomized to FFF (eight weekly 2 hour sessions) or the control condition (a fall prevention brochure and informing their neurologist of their fall history). Participants counted falls prospectively from enrollment through 6 months following intervention. Effects on fall frequency were evaluated by the Bayesian analysis.

RESULTS

The modeled mean fall frequency pre-intervention was 1.2 falls/month in the FFF group (95% credible intervals (CIs) = 0.8-2.0) and 1.4 falls/month in the control group (95% CI = 0.9-2.1). Fall frequency decreased by 0.6 falls/month in both groups over time (nadir 4-6 months post-intervention: FFF 0.6 falls/month (95% CI = 0.4-0.9); control 0.8 falls/month (95% CI = 0.5-1.1)).

CONCLUSION

In-person group exercise and education are not superior to written education and neurologist-initiated interventions for preventing falls in PwMS.

Comparing fall detection methods in people with multiple sclerosis: A prospective observational cohort study

Background Falls occur across the population but are more common, and have more negative sequelae, in people with multiple sclerosis (MS). Given the prevalence and impact of falls, accurate measures of fall frequency are needed. This study compares the sensitivity and false discovery rates of three methods of fall detection: the current gold standard, prospective paper fall calendars, real-time self-reporting and automated detection, the latter two from a novel body-worn device. Methods Falls in twenty-five people with MS were recorded for eight weeks with prospective fall calendars, real-time body-worn self-report, and an automated body-worn detector concurrently. Eligible individuals were adults with MS enrolled in a randomized controlled trial of a fall prevention intervention. Entry criteria were at least two falls or near-falls in the previous two months, Expanded Disability Status Scale ≤ 6.0, community dwelling, and no MS relapse in the previous month. The sensitivity (proportion of true falls detected) and false discovery rates (proportion of false reports generated) of the fall detection methods were compared. A true fall was a fall reported by at least two methods. A false report was a fall reported by only one method. The trial is registered on ClinicalTrials.gov (NCT02583386) and is closed. Results In the 1,276 person-days of fall counting with all three methods in use simultaneously there were 1344 unique fall events. Of these, 8.5% (114) were true falls and 91.5% (1230) were false reports. Fall calendars had the lowest sensitivity (0.614) and the lowest false discovery rate (0.067). The automated detector had the highest sensitivity (0.921) and the highest false discovery rate (0.919). All methods generated under one false report per day. There were no fall detection-related adverse events. Conclusion Fall calendars likely underestimate fall frequency by around 40%. The automated detector evaluated here misses very few falls but likely overestimates the number of falls by around one fall per day. Additional research is needed to produce an ideal fall detection and counting method for use in clinical and research applications. Funding United States Department of Veterans Affairs, Rehabilitations Research and Development Service.

Automated Detection of Real-World Falls: Modeled From People With Multiple Sclerosis

Falls are a major health problem with one in three people over the age of 65 falling each year, oftentimes causing hip fractures, disability, reduced mobility, hospitalization and death. A major limitation in fall detection algorithm development is an absence of real-world falls data. Fall detection algorithms are typically trained on simulated fall data that contain a well-balanced number of examples of falls and activities of daily living. However, real-world falls occur infrequently, making them difficult to capture and causing severe data imbalance. People with multiple sclerosis (MS) fall frequently, and their risk of falling increases with disease progression. Because of their high fall incidence, people with MS provide an ideal model for studying falls. This paper describes the development of a context-aware fall detection system based on inertial sensors and time of flight sensors that is robust to imbalance, which is trained and evaluated on real-world falls in people with MS. The algorithm uses an auto-encoder that detects fall candidates using reconstruction error of accelerometer signals followed by a hyper-ensemble of balanced random forests trained using both acceleration and movement features. On a clinical dataset obtained from 25 people with MS monitored over eight weeks during free-living conditions, 54 falls were observed and our system achieved a sensitivity of 92.14%, and false-positive rate of 0.65 false alarms per day.

More Time in Glucose Range During Exercise Days than Sedentary Days in Adults Living with Type 1 Diabetes

Objective: This study analysis was designed to examine the 24-h effects of exercise on glycemic control as measured by continuous glucose monitoring (CGM). Methods: Individuals with type 1 diabetes (ages: 15-68 years; hemoglobin A1c: 7.5% ± 1.5% [mean ± standard deviation (SD)]) were randomly assigned to complete twice-weekly aerobic, high-intensity interval, or resistance-based exercise sessions in addition to their personal exercise sessions for a period of 4 weeks. Exercise was tracked with wearables and glucose concentrations assessed using CGM. An exercise day was defined as a 24-h period after the end of exercise, while a sedentary day was defined as any 24-h period with no recorded exercise ≥10 min long. Sedentary days start at least 24 h after the end of exercise. Results: Mean glucose was lower (150 ± 45 vs. 166 ± 49 mg/dL, P = 0.01), % time in range [70-180 mg/dL] higher (62% ± 23% vs. 56% ± 25%, P = 0.03), % time >180 mg/dL lower (28% ± 23% vs. 37% ± 26%, P = 0.01), and % time <70 mg/dL higher (9.3% ± 11.0% vs. 7.1% ± 9.1%, P = 0.04) on exercise days compared with sedentary days. Glucose variability and % time <54 mg/dL did not differ significantly between exercise and sedentary days. No significant differences in glucose control by exercise type were observed. Conclusion: Participants had lower 24-h mean glucose levels and a greater time in range on exercise days compared with sedentary days, with mode of exercise affecting glycemia similarly. In summary, this study offers data supporting frequency of exercise as a method of facilitating glucose control but does not suggest an effect for mode of exercise.

Separating insulin-mediated and non-insulin-mediated glucose uptake during and after aerobic exercise in type 1 diabetes

Aerobic exercise in type 1 diabetes (T1D) causes rapid increase in glucose utilization due to muscle work during exercise, followed by increased insulin sensitivity after exercise. Better understanding of these changes is necessary for models of exercise in T1D. Twenty-six individuals with T1D underwent three sessions at three insulin rates (100%, 150%, 300% of basal). After 3-h run-in, participants performed 45 min aerobic exercise (moderate or intense). We determined area under the curve for endogenous glucose production (AUC_EGP) and rate of glucose disappearance (AUC_Rd) over 45 min from exercise start. A novel application of linear regression of R_d across the three insulin sessions allowed separation of insulin-mediated from non-insulin-mediated glucose uptake before, during, and after exercise. AUC_Rd increased 12.45 mmol/L (CI = 10.33-14.58, P < 0.001) and 13.13 mmol/L (CI = 11.01-15.26, P < 0.001) whereas AUC_EGP increased 1.66 mmol/L (CI = 1.01-2.31, P < 0.001) and 3.46 mmol/L (CI = 2.81-4.11, P < 0.001) above baseline during moderate and intense exercise, respectively. AUC_EGP increased during intense exercise by 2.14 mmol/L (CI = 0.91-3.37, P < 0.001) compared with moderate exercise. There was significant effect of insulin infusion rate on AUC_Rd equal to 0.06 mmol/L per % above basal rate (CI = 0.05-0.07, P < 0.001). Insulin-mediated glucose uptake rose during exercise and persisted hours afterward, whereas non-insulin-mediated effect was limited to the exercise period. To our knowledge, this method of isolating dynamic insulin- and non-insulin-mediated uptake has not been previously employed during exercise. These results will be useful in informing glucoregulatory models of T1D. The study has been registered at www.clinicaltrials.gov as NCT03090451.NEW & NOTEWORTHY Separating insulin and non-insulin glucose uptake dynamically during exercise in type 1 diabetes has not been done before. We use a multistep process, including a previously described linear regression method, over three insulin infusion sessions, to perform this separation and can graph these components before, during, and after exercise for the first time.

Comparison Between Closed-Loop Insulin Delivery System (the Artificial Pancreas) and Sensor-Augmented Pump Therapy: A Randomized-Controlled Crossover Trial

Objective: Several studies have shown that closed-loop automated insulin delivery (the artificial pancreas) improves glucose control compared with sensor-augmented pump therapy. We aimed to confirm these findings using our automated insulin delivery system based on the iPancreas platform. Research Design and Methods: We conducted a two-center, randomized crossover trial comparing automated insulin delivery with sensor-augmented pump therapy in 36 adults with type 1 diabetes. Each intervention lasted 12 days in outpatient free-living conditions with no remote monitoring. The automated insulin delivery system used a model predictive control algorithm that was a less aggressive version of our earlier dosing algorithm to emphasize safety. The primary outcome was time in the range 3.9-10.0 mmol/L. Results: The automated insulin delivery system was operational 90.2% of the time. Compared with the sensor-augmented pump therapy, automated insulin delivery increased time in range (3.9-10.0 mmol/L) from 61% (interquartile range 53-74) to 69% (60-73; P = 0.006) and increased time in tight target range (3.9-7.8 mmol/L) from 37% (30-49) to 45% (35-51; P = 0.011). Automated insulin delivery also reduced time spent below 3.9 and 3.3 mmol/L from 3.5% (0.8-5.4) to 1.6% (1.1-2.7; P = 0.0021) and from 0.9% (0.2-2.1) to 0.5% (0.2-1.1; P = 0.0122), respectively. Time spent below 2.8 mmol/L was 0.2% (0.0-0.6) with sensor-augmented pump therapy and 0.1% (0.0-0.4; P = 0.155) with automated insulin delivery. Conclusions: Our study confirms findings that automated insulin delivery improves glucose control compared with sensor-augmented pump therapy. ClinicalTrials.gov no. NCT02846831.

Prediction of Mild Cognitive Impairment Using Movement Complexity

OBJECTIVE

Aimless movement or wandering may be a symptom of mild cognitive impairment (MCI) that arises as a consequence of confusion and forgetfulness. This paper presents a support vector machine (SVM) framework based on movement analysis for the prediction of the onset and progression of MCI.

METHODS

Movement data of 22 subjects with MCI, and 22 other healthy subjects, living independently in smart homes were collected for ten years using motion sensors. Features were extracted from the sensor data using movement metrics, including cyclomatic complexity, detrended fluctuation analysis, fractal index, entropy, and room transitions. Two different SVM classification algorithms were trained using the features, first to predict the progression of MCI in the post-transition period, and second to predict the onset of MCI in the pre-transition phase.

RESULTS

The two SVMs were able to detect the onset six months earlier than the clinical diagnosis. The model accuracy in classifying MCI increased monotonically from the onset month and reached maximum (81%) at the 11^th post-transition month. The features of cyclomatic complexity contributed significantly to the prediction results.

CONCLUSION

Findings support the use of movement complexity measures and machine learning for monitoring cognitive behavior in an independent living environment.

Glucose management for exercise using continuous glucose monitoring (CGM) and intermittently scanned CGM (isCGM) systems in type 1 diabetes: position statement of the European Association for the Study of Diabetes (EASD) and of the International Society for Pediatric and Adolescent Diabetes (ISPAD) endorsed by JDRF and supported by the American Diabetes Association (ADA)

Physical exercise is an important component in the management of type 1 diabetes across the lifespan. Yet, acute exercise increases the risk of dysglycaemia, and the direction of glycaemic excursions depends, to some extent, on the intensity and duration of the type of exercise. Understandably, fear of hypoglycaemia is one of the strongest barriers to incorporating exercise into daily life. Risk of hypoglycaemia during and after exercise can be lowered when insulin-dose adjustments are made and/or additional carbohydrates are consumed. Glycaemic management during exercise has been made easier with continuous glucose monitoring (CGM) and intermittently scanned continuous glucose monitoring (isCGM) systems; however, because of the complexity of CGM and isCGM systems, both individuals with type 1 diabetes and their healthcare professionals may struggle with the interpretation of given information to maximise the technological potential for effective use around exercise (i.e. before, during and after). This position statement highlights the recent advancements in CGM and isCGM technology, with a focus on the evidence base for their efficacy to sense glucose around exercise and adaptations in the use of these emerging tools, and updates the guidance for exercise in adults, children and adolescents with type 1 diabetes. Graphical abstract.

Glucose management for exercise using continuous glucose monitoring (CGM) and intermittently scanned CGM (isCGM) systems in type 1 diabetes: position statement of the European Association for the Study of Diabetes (EASD) and of the International Society for Pediatric and Adolescent Diabetes (ISPAD) endorsed by JDRF and supported by the American Diabetes Association (ADA)

Physical exercise is an important component in the management of type 1 diabetes across the lifespan. Yet, acute exercise increases the risk of dysglycaemia, and the direction of glycaemic excursions depends, to some extent, on the intensity and duration of the type of exercise. Understandably, fear of hypoglycaemia is one of the strongest barriers to incorporating exercise into daily life. Risk of hypoglycaemia during and after exercise can be lowered when insulin-dose adjustments are made and/or additional carbohydrates are consumed. Glycaemic management during exercise has been made easier with continuous glucose monitoring (CGM) and intermittently scanned continuous glucose monitoring (isCGM) systems; however, because of the complexity of CGM and isCGM systems, both individuals with type 1 diabetes and their healthcare professionals may struggle with the interpretation of given information to maximise the technological potential for effective use around exercise (ie, before, during and after). This position statement highlights the recent advancements in CGM and isCGM technology, with a focus on the evidence base for their efficacy to sense glucose around exercise and adaptations in the use of these emerging tools, and updates the guidance for exercise in adults, children and adolescents with type 1 diabetes.

Predicting and Preventing Nocturnal Hypoglycemia in Type 1 Diabetes Using Big Data Analytics and Decision Theoretic Analysis

Background: Despite new glucose sensing technologies, nocturnal hypoglycemia is still a problem for people with type 1 diabetes (T1D) as symptoms and sensor alarms may not be detected while sleeping. Accurately predicting nocturnal hypoglycemia before sleep may help minimize nighttime hypoglycemia. Methods: A support vector regression (SVR) model was trained to predict, before bedtime, the overnight minimum glucose and overnight nocturnal hypoglycemia for people with T1D. The algorithm was trained on continuous glucose measurements and insulin data collected from 124 people (22,804 valid nights of data) with T1D. The minimum glucose threshold for announcing nocturnal hypoglycemia risk was derived by applying a decision theoretic criterion to maximize expected net benefit. Accuracy was evaluated on a validation set from 10 people with T1D during a 4-week trial under free-living sensor-augmented insulin-pump therapy. The primary outcome measures were sensitivity and specificity of prediction, the correlation between predicted and actual minimum nocturnal glucose, and root-mean-square error. The impact of using the algorithm to prevent nocturnal hypoglycemia is shown in-silico. Results: The algorithm predicted 94.1% of nocturnal hypoglycemia events (<3.9 mmol/L, 95% confidence interval [CI], 71.3-99.9) with an area under the receiver operating characteristic curve of 0.86 (95% CI, 0.75-0.98). Correlation between actual and predicted minimum glucose was high (R = 0.71, P < 0.001). In-silico simulations showed that the algorithm could reduce nocturnal hypoglycemia by 77.0% (P = 0.006) without impacting time in target range (3.9-10 mmol/L). Conclusion: An SVR model trained on a big data set and optimized using decision theoretic criterion can accurately predict at bedtime if overnight nocturnal hypoglycemia will occur and may help reduce nocturnal hypoglycemia.

Dual-Hormone Closed-Loop System Using a Liquid Stable Glucagon Formulation Versus Insulin-Only Closed-Loop System Compared With a Predictive Low Glucose Suspend System: An Open-Label, Outpatient, Single-Center, Crossover, Randomized Controlled Trial

OBJECTIVE

To assess the efficacy and feasibility of a dual-hormone (DH) closed-loop system with insulin and a novel liquid stable glucagon formulation compared with an insulin-only closed-loop system and a predictive low glucose suspend (PLGS) system.

RESEARCH DESIGN AND METHODS

In a 76-h, randomized, crossover, outpatient study, 23 participants with type 1 diabetes used three modes of the Oregon Artificial Pancreas system: 1) dual-hormone (DH) closed-loop control, 2) insulin-only single-hormone (SH) closed-loop control, and 3) PLGS system. The primary end point was percentage time in hypoglycemia (<70 mg/dL) from the start of in-clinic aerobic exercise (45 min at 60% VO_2max) to 4 h after.

RESULTS

DH reduced hypoglycemia compared with SH during and after exercise (DH 0.0% [interquartile range 0.0-4.2], SH 8.3% [0.0-12.5], P = 0.025). There was an increased time in hyperglycemia (>180 mg/dL) during and after exercise for DH versus SH (20.8% DH vs. 6.3% SH, P = 0.038). Mean glucose during the entire study duration was DH, 159.2; SH, 151.6; and PLGS, 163.6 mg/dL. Across the entire study duration, DH resulted in 7.5% more time in target range (70-180 mg/dL) compared with the PLGS system (71.0% vs. 63.4%, P = 0.044). For the entire study duration, DH had 28.2% time in hyperglycemia vs. 25.1% for SH (P = 0.044) and 34.7% for PLGS (P = 0.140). Four participants experienced nausea related to glucagon, leading three to withdraw from the study.

CONCLUSIONS

The glucagon formulation demonstrated feasibility in a closed-loop system. The DH system reduced hypoglycemia during and after exercise, with some increase in hyperglycemia.

Patient Input for Design of a Decision Support Smartphone Application for Type 1 Diabetes

BACKGROUND

Decision support smartphone applications integrated with continuous glucose monitors may improve glycemic control in type 1 diabetes (T1D). We conducted a survey to understand trends and needs of potential users to inform the design of decision support technology.

METHODS

A 70-question survey was distributed October 2017 through May 2018 to adults aged 18-80 with T1D from a specialty clinic and T1D Exchange online health community (myglu.org). The survey responses were used to evaluate potential features of a diabetes decision support tool by Likert scale and open responses.

RESULTS

There were 1542 responses (mean age 46.1 years [SD 15.2], mean duration of diabetes 26.5 years [SD 15.8]). The majority (84.2%) have never used an app to manage diabetes; however, a large majority (77.8%) expressed interest in using a decision support app. The ability to predict and avoid hypoglycemia was the most important feature identified by a majority of the respondents, with 91% of respondents indicating the highest level of interest in these features. The task that respondents find most difficult was management of glucose during exercise (only 47% of participants were confident in glucose management during exercise). The respondents also highly desired features that help manage glucose during exercise (85% of respondents were interested). The responses identified integration and interoperability with peripheral devices/apps and customization of alerts as important. Responses from participants were generally consistent across stratified categories.

CONCLUSIONS

These results provide valuable insight into patient needs in decision support applications for management of T1D.

Accuracy of the Dexcom G6 Glucose Sensor during Aerobic, Resistance, and Interval Exercise in Adults with Type 1 Diabetes

The accuracy of continuous glucose monitoring (CGM) sensors may be significantly impacted by exercise. We evaluated the impact of three different types of exercise on the accuracy of the Dexcom G6 sensor. Twenty-four adults with type 1 diabetes on multiple daily injections wore a G6 sensor. Participants were randomized to aerobic, resistance, or high intensity interval training (HIIT) exercise. Each participant completed two in-clinic 30-min exercise sessions. The sensors were applied on average 5.3 days prior to the in-clinic visits (range 0.6–9.9). Capillary blood glucose (CBG) measurements with a Contour Next meter were performed before and after exercise as well as every 10 min during exercise. No CGM calibrations were performed. The median absolute relative difference (MARD) and median relative difference (MRD) of the CGM as compared with the reference CBG did not differ significantly from the start of exercise to the end exercise across all exercise types (ranges for aerobic MARD: 8.9 to 13.9% and MRD: −6.4 to 0.5%, resistance MARD: 7.7 to 14.5% and MRD: −8.3 to −2.9%, HIIT MARD: 12.1 to 16.8% and MRD: −14.3 to −9.1%). The accuracy of the no-calibration Dexcom G6 CGM was not significantly impacted by aerobic, resistance, or HIIT exercise.

An artificial intelligence decision support system for the management of type 1 diabetes

Type 1 diabetes (T1D) is characterized by pancreatic beta cell dysfunction and insulin depletion. Over 40% of people with T1D manage their glucose through multiple injections of long-acting basal and short-acting bolus insulin, so-called multiple daily injections (MDI)^1,2. Errors in dosing can lead to life-threatening hypoglycaemia events (<70 mg dl^-1) and hyperglycaemia (>180 mg dl^-1), increasing the risk of retinopathy, neuropathy, and nephropathy. Machine learning (artificial intelligence) approaches are being harnessed to incorporate decision support into many medical specialties. Here, we report an algorithm that provides weekly insulin dosage recommendations to adults with T1D using MDI therapy. We employ a unique virtual platform³ to generate over 50,000 glucose observations to train a k-nearest neighbours⁴ decision support system (KNN-DSS) to identify causes of hyperglycaemia or hypoglycaemia and determine necessary insulin adjustments from a set of 12 potential recommendations. The KNN-DSS algorithm achieves an overall agreement with board-certified endocrinologists of 67.9% when validated on real-world human data, and delivers safe recommendations, per endocrinologist review. A comparison of inter-physician-recommended adjustments to insulin pump therapy indicates full agreement of 41.2% among endocrinologists, which is consistent with previous measures of inter-physician agreement (41-45%)⁵. In silico^3,6 benchmarking using a platform accepted by the United States Food and Drug Administration for evaluation of artificial pancreas technologies indicates substantial improvement in glycaemic outcomes after 12 weeks of KNN-DSS use. Our data indicate that the KNN-DSS allows for early identification of dangerous insulin regimens and may be used to improve glycaemic outcomes and prevent life-threatening complications in people with T1D.

Artificial Intelligence in Decision Support Systems for Type 1 Diabetes

Type 1 diabetes (T1D) is a chronic health condition resulting from pancreatic beta cell dysfunction and insulin depletion. While automated insulin delivery systems are now available, many people choose to manage insulin delivery manually through insulin pumps or through multiple daily injections. Frequent insulin titrations are needed to adequately manage glucose, however, provider adjustments are typically made every several months. Recent automated decision support systems incorporate artificial intelligence algorithms to deliver personalized recommendations regarding insulin doses and daily behaviors. This paper presents a comprehensive review of computational and artificial intelligence-based decision support systems to manage T1D. Articles were obtained from PubMed, IEEE Xplore, and ScienceDirect databases. No time period restrictions were imposed on the search. After removing off-topic articles and duplicates, 562 articles were left to review. Of those articles, we identified 61 articles for comprehensive review based on algorithm evaluation using real-world human data, in silico trials, or clinical studies. We grouped decision support systems into general categories of (1) those which recommend adjustments to insulin and (2) those which predict and help avoid hypoglycemia. We review the artificial intelligence methods used for each type of decision support system, and discuss the performance and potential applications of these systems.

Role of Glucagon in Automated Insulin Delivery

Treatment of type 1 diabetes with exogenous insulin often results in unpredictable daily glucose variability and hypoglycemia, which can be dangerous. Automated insulin delivery systems can improve glucose control while reducing burden for people with diabetes. One approach to improve treatment outcomes is to incorporate the counter-regulatory hormone glucagon into the automated delivery system to help prevent the hypoglycemia that can be induced by the slow pharmacodynamics of insulin action. This article explores the advantages and disadvantages of incorporating glucagon into dual-hormone automated hormone delivery systems.

Adaptive Control of an Artificial Pancreas Using Model Identification, Adaptive Postprandial Insulin Delivery, and Heart Rate and Accelerometry as Control Inputs

BACKGROUND

People with type 1 diabetes (T1D) have varying sensitivities to insulin and also varying responses to meals and exercise. We introduce a new adaptive run-to-run model predictive control (MPC) algorithm that can be used to help people with T1D better manage their glucose levels using an artificial pancreas (AP). The algorithm adapts to individuals' different insulin sensitivities, glycemic response to meals, and adjustment during exercise as a continuous input during free-living conditions.

METHODS

A new insulin sensitivity adaptation (ISA) algorithm is presented that updates each patient's insulin sensitivity during nonmeal periods to reduce the error between the actual glucose levels and the process model. We further demonstrate how an adaptive learning postprandial hypoglycemia prevention algorithm (ALPHA) presented in the previous work can complement the ISA algorithm, and the algorithm can adapt in several days. We further show that if physical activity is incorporated as a continuous input (heart rate and accelerometry), performance is improved. The contribution of this work is the description of the ISA algorithm and the evaluation of how ISA, ALPHA, and incorporation of exercise metrics as a continuous input can impact glycemic control.

RESULTS

Incorporating ALPHA, ISA, and physical activity into the MPC improved glycemic outcome measures. The adaptive learning postprandial hypoglycemia prevention algorithm combined with ISA significantly reduced time spent in hypoglycemia by 71.7% and the total number of rescue carbs by 67.8% to 0.37% events/day/patient. Insulin sensitivity adaptation significantly reduced model-actual mismatch by 12.2% compared to an AP without ISA. Incorporating physical activity as a continuous input modestly improved time in the range 70 to 180 mg/dL during high physical activity days from 84.4% to 84.9% and reduced the percentage time in hypoglycemia by 23.8% from 2.1% to 1.6%.

CONCLUSION

Adapting postprandial insulin delivery, insulin sensitivity, and adapting to physical exercise in an MPC-based AP systems can improve glycemic outcomes.

Prediction of Hypoglycemia During Aerobic Exercise in Adults With Type 1 Diabetes

BACKGROUND

Fear of exercise related hypoglycemia is a major reason why people with type 1 diabetes (T1D) do not exercise. There is no validated prediction algorithm that can predict hypoglycemia at the start of aerobic exercise.

METHODS

We have developed and evaluated two separate algorithms to predict hypoglycemia at the start of exercise. Model 1 is a decision tree and model 2 is a random forest model. Both models were trained using a meta-data set based on 154 observations of in-clinic aerobic exercise in 43 adults with T1D from 3 different studies that included participants using sensor augmented pump therapy, automated insulin delivery therapy, and automated insulin and glucagon therapy. Both models were validated using an entirely new validation data set with 90 exercise observations collected from 12 new adults with T1D.

RESULTS

Model 1 identified two critical features predictive of hypoglycemia during exercise: heart rate and glucose at the start of exercise. If heart rate was greater than 121 bpm during the first 5 min of exercise and glucose at the start of exercise was less than 182 mg/dL, it predicted hypoglycemia with 79.55% accuracy. Model 2 achieved a higher accuracy of 86.7% using additional features and higher complexity.

CONCLUSIONS

Models presented here can assist people with T1D to avoid exercise related hypoglycemia. The simple model 1 heuristic can be easily remembered (the 180/120 rule) and model 2 is more complex requiring computational resources, making it suitable for automated artificial pancreas or decision support systems.

A statistical virtual patient population for the glucoregulatory system in type 1 diabetes with integrated exercise model

Purpose

We introduce two validated single (SH) and dual hormone (DH) mathematical models that represent an in-silico virtual patient population (VPP) for type 1 diabetes (T1D). The VPP can be used to evaluate automated insulin and glucagon delivery algorithms, so-called artificial pancreas (AP) algorithms that are currently being used to help people with T1D better manage their glucose levels. We present validation results comparing these virtual patients with true clinical patients undergoing AP control and demonstrate that the virtual patients behave similarly to people with T1D.

Methods

A single hormone virtual patient population (SH-VPP) was created that is comprised of eight differential equations that describe insulin kinetics, insulin dynamics and carbohydrate absorption. The parameters in this model that represent insulin sensitivity were statistically sampled from a normal distribution to create a population of virtual patients with different levels of insulin sensitivity. A dual hormone virtual patient population (DH-VPP) extended this SH-VPP by incorporating additional equations to represent glucagon kinetics and glucagon dynamics. The DH-VPP is comprised of thirteen differential equations and a parameter representing glucagon sensitivity, which was statistically sampled from a normal distribution to create virtual patients with different levels of glucagon sensitivity. We evaluated the SH-VPP and DH-VPP on a clinical data set of 20 people with T1D who participated in a 3.5-day outpatient AP study. Twenty virtual patients were matched with the 20 clinical patients by total daily insulin requirements and body weight. The identical meals given during the AP study were given to the virtual patients and the identical AP control algorithm that was used to control the glucose of the virtual patients was used on the clinical patients. We compared percent time in target range (70–180 mg/dL), time in hypoglycemia (<70 mg/dL) and time in hyperglycemia (>180 mg/dL) for both the virtual patients and the actual patients.

Results

The subjects in the SH-VPP performed similarly vs. the actual patients (time in range: 78.1 ± 5.1% vs. 74.3 ± 8.1%, p = 0.11; time in hypoglycemia: 3.4 ± 1.3% vs. 2.8 ± 1.7%, p = 0.23). The subjects in the DH-VPP also performed similarly vs. the actual patients (time in range: 75.6 ± 5.5% vs. 71.9 ± 10.9%, p = 0.13; time in hypoglycemia: 0.9 ± 0.8% vs. 1.3 ± 1%, p = 0.19). While the VPPs tended to over-estimate the time in range relative to actual patients, the difference was not statistically significant.

Conclusions

We have verified that a SH-VPP and a DH-VPP performed comparably with actual patients undergoing AP control using an identical control algorithm. The SH-VPP and DH-VPP may be used as a simulator for pre-evaluation of T1D control algorithms.

Design and evaluation of a portable smart-phone based peripheral neuropathy test platform

In this paper, we describe a novel portable test platform that can be used to test peripheral neuropathy either within a clinic or at home. The system, called the PeriVib, is comprised of (1) a small, custom vibration motor designed to apply a vibration stimulus to the toe with constant pressure to test sensation threshold, and (2) a custom smart-phone app that enables a patient to run a series of functional gait and balance tests. Vibration is applied by PeriVib in two separate modes. The first mode, ramp-up, starts at zero amplitude and increases to a maximum level while the patient indicates when they start feeling the pressure by lifting their finger off the touch-screen on the phone. The second mode, ramp-down, starts at a maximal intensity and decreases in intensity; the patient indicates when they stop feeling the vibration. The smart-phone app determines the patient's threshold by recording the vibration amplitude when they indicate the onset or loss of vibratory sensation, depending on the mode. In both modes, the measurement is repeated five times. In addition to controlling the vibration motor during the vibration test, the smart phone app also enables collection of gait and sway metrics through the use of the accelerometer and gyroscope sensors on the smartphone. The entire set of tests requires approximately 5 minutes to complete and can be done by a patient with minimal instructions from a clinician. In a cohort of 28 subjects with a history of chemotherapy-induced peripheral neuropathy, we compared the PeriVib performance with two established threshold sensing systems: (1) a Biothesiometer device and (2) a tuning fork. We found that the sensation threshold estimated by PeriVib correlated well with the Biothesiometer ($\mathrm{R}^{2}$ of 0.68) but less well with the tuning fork ($\mathrm{R}^{2}$ of 0.15). Functional gait and balance metrics did not correlate with peripheral neuropathy severity.

Accuracy of Wrist-Worn Activity Monitors During Common Daily Physical Activities and Types of Structured Exercise: Evaluation Study

BACKGROUND

Wrist-worn activity monitors are often used to monitor heart rate (HR) and energy expenditure (EE) in a variety of settings including more recently in medical applications. The use of real-time physiological signals to inform medical systems including drug delivery systems and decision support systems will depend on the accuracy of the signals being measured, including accuracy of HR and EE. Prior studies assessed accuracy of wearables only during steady-state aerobic exercise.

OBJECTIVE

The objective of this study was to validate the accuracy of both HR and EE for 2 common wrist-worn devices during a variety of dynamic activities that represent various physical activities associated with daily living including structured exercise.

METHODS

We assessed the accuracy of both HR and EE for two common wrist-worn devices (Fitbit Charge 2 and Garmin vívosmart HR+) during dynamic activities. Over a 2-day period, 20 healthy adults (age: mean 27.5 [SD 6.0] years; body mass index: mean 22.5 [SD 2.3] kg/m²; 11 females) performed a maximal oxygen uptake test, free-weight resistance circuit, interval training session, and activities of daily living. Validity was assessed using an HR chest strap (Polar) and portable indirect calorimetry (Cosmed). Accuracy of the commercial wearables versus research-grade standards was determined using Bland-Altman analysis, correlational analysis, and error bias.

RESULTS

Fitbit and Garmin were reasonably accurate at measuring HR but with an overall negative bias. There was more error observed during high-intensity activities when there was a lack of repetitive wrist motion and when the exercise mode indicator was not used. The Garmin estimated HR with a mean relative error (RE, %) of -3.3% (SD 16.7), whereas Fitbit estimated HR with an RE of -4.7% (SD 19.6) across all activities. The highest error was observed during high-intensity intervals on bike (Fitbit: -11.4% [SD 35.7]; Garmin: -14.3% [SD 20.5]) and lowest error during high-intensity intervals on treadmill (Fitbit: -1.7% [SD 11.5]; Garmin: -0.5% [SD 9.4]). Fitbit and Garmin EE estimates differed significantly, with Garmin having less negative bias (Fitbit: -19.3% [SD 28.9], Garmin: -1.6% [SD 30.6], P<.001) across all activities, and with both correlating poorly with indirect calorimetry measures.

CONCLUSIONS

Two common wrist-worn devices (Fitbit Charge 2 and Garmin vívosmart HR+) show good HR accuracy, with a small negative bias, and reasonable EE estimates during low to moderate-intensity exercise and during a variety of common daily activities and exercise. Accuracy was compromised markedly when the activity indicator was not used on the watch or when activities involving less wrist motion such as cycle ergometry were done.

In-Home Sleep Apnea Severity Classification using Contact-free Load Cells and an AdaBoosted Decision Tree Algorithm

We present a method for automated diagnosis and classification of severity of sleep apnea using an array of non-contact pressure-sensitive sensors placed underneath a mattress as an alternative to conventional obtrusive sensors. Our algorithm comprises two stages: i) A decision tree classifier that identifies patients with sleep apnea, and ii) a subsequent linear regression model that estimates the Apnea-Hypopnea Index (AHI), which is used to determine the severity of sleep disordered breathing. We tested our algorithm on a cohort of 14 patients who underwent overnight home sleep apnea test. The machine learning algorithm was trained and performance was evaluated using leave-one-patient-out cross-validation. The accuracy of the proposed approach in detecting sleep apnea is 86.96%, with sensitivity and specificity of 81.82% and 91.67%, respectively. Moreover, classification of severity of the sleep disorder was correctly assigned in 11 out of 14 cases, and the mean absolute error in the AHI estimation was calculated to be 3.83 events/hr.

Effect of Aerobic and Resistance Exercise on Glycemic Control in Adults With Type 1 Diabetes

OBJECTIVES

Physical exercise is recommended for individuals with type 1 diabetes, yet the effects of exercise on glycemic control are not well established. We evaluated the impact of different modes of exercise on glycemic control in people with type 1 diabetes.

METHODS

In a 3-week randomized crossover trial, 10 adults with type 1 diabetes (4 men and 6 women, aged 33±6 years; duration of diabetes, 18±10 years; glycated hemoglobin level, 7.4%±1%) were assigned to 3 weeks of intervention: aerobic exercise (treadmill at 60% of maximum volume of oxygen utilization), resistance training (8 to 12 repetitions of 5 upper and lower body exercises at 60% to 80% of 1 repetition maximum) or no exercise (control). During each exercise week, participants completed 2 monitored 45 min exercise sessions. For each week of the study, we analyzed participants' insulin pump data, sensor glucose data and meal intake using a custom smart-phone application. The primary outcome was the percentage of time in range (glucose >3.9 mmol/L and ≤10 mmol/L) for the 24 h after each bout of exercise or rest during the control week. The study was registered on ClinicalTrials.gov (NCT:02687893).

RESULTS

Aerobic exercise caused a mean glucose reduction during exercise of 3.94±2.67 mmol/L, whereas the reduction during resistance training was 1.33±1.78 mmol/L (p=0.007). The mean percentage time in range for the 24 h after resistance training was significantly greater than that during the control period (70% vs. 56%, p=0.013) but not after aerobic exercise (60%).

CONCLUSIONS

The results indicate that when various confounders are considered, resistance training could improve glycemic control in this population.

Randomized Outpatient Trial of Single- and Dual-Hormone Closed-Loop Systems That Adapt to Exercise Using Wearable Sensors

OBJECTIVE

Automated insulin delivery is the new standard for type 1 diabetes, but exercise-related hypoglycemia remains a challenge. Our aim was to determine whether a dual-hormone closed-loop system using wearable sensors to detect exercise and adjust dosing to reduce exercise-related hypoglycemia would outperform other forms of closed-loop and open-loop therapy.

RESEARCH DESIGN AND METHODS

Participants underwent four arms in randomized order: dual-hormone, single-hormone, predictive low glucose suspend, and continuation of current care over 4 outpatient days. Each arm included three moderate-intensity aerobic exercise sessions. The two primary outcomes were percentage of time in hypoglycemia (<70 mg/dL) and in a target range (70-180 mg/dL) assessed across the entire study and from the start of the in-clinic exercise until the next meal.

RESULTS

The analysis included 20 adults with type 1 diabetes who completed all arms. The mean time (SD) in hypoglycemia was the lowest with dual-hormone during the exercise period: 3.4% (4.5) vs. 8.3% (12.6) single-hormone (P = 0.009) vs. 7.6% (8.0) predictive low glucose suspend (P < 0.001) vs. 4.3% (6.8) current care where pre-exercise insulin adjustments were allowed (P = 0.49). Time in hypoglycemia was also the lowest with dual-hormone during the entire 4-day study: 1.3% (1.0) vs. 2.8% (1.7) single-hormone (P < 0.001) vs. 2.0% (1.5) predictive low glucose suspend (P = 0.04) vs. 3.1% (3.2) current care (P = 0.007). Time in range during the entire study was the highest with single-hormone: 74.3% (8.0) vs. 72.0% (10.8) dual-hormone (P = 0.44).

CONCLUSIONS

The addition of glucagon delivery to a closed-loop system with automated exercise detection reduces hypoglycemia in physically active adults with type 1 diabetes.

The effect of exercise on sleep in adults with type 1 diabetes

The aim of this pilot study was to investigate the effect of exercise on sleep and nocturnal hypoglycaemia in adults with type 1 diabetes (T1D). In a 3-week crossover trial, 10 adults with T1D were randomized to perform aerobic, resistance or no exercise. During each exercise week, participants completed 2 separate 45-minutes exercise sessions at an academic medical center. Participants returned home and wore a continuous glucose monitor and a wrist-based activity monitor to estimate sleep duration. Participants on average lost 70 (±49) minutes of sleep (P = .0015) on nights following aerobic exercise and 27 (±78) minutes (P = .3) following resistance exercise relative to control nights. The odds ratio with confidence intervals of nocturnal hypoglycaemia occurring on nights following aerobic and resistance exercise was 5.4 (1.3, 27.2) and 7.0 (1.7, 37.3), respectively. Aerobic exercise can cause sleep loss in T1D possibly from increased hypoglycaemia.

Ubiquitous Real-World Sensing and Audiology-Based Health Informatics

BACKGROUND

Hearing impairment and hearing rehabilitation strategies have historically been studied within the confines of a sound booth under controlled experimental conditions. The real world is quite different from the clinical setting and it is important to study how a person with hearing impairment interacts with the world both with and without a hearing assist intervention. A person's ability to hear enables them to communicate and to effectively interact with the world. If a person suffers from hearing impairment, we might anticipate that they could become more disengaged from the world, more socially isolated, potentially depressed, and have additional comorbidities such as cognitive and physical impairment. Indeed, prior research has shown that hearing impairment is associated with social isolation, decreased functional ability and mobility, fall risk, diabetes, and cognitive impairment. However, nearly all of the work that has been done in this area of assessing the impact of hearing impairment on a person's social, cognitive, and physical health has been done through clinical tests or self-report studies using questionnaires and surveys that attempt to objectively quantify various aspects of health. Unfortunately, clinical tests, questionnaires, and surveys oftentimes inaccurately assess a person's true social, cognitive, and physical health. Only when a person is observed in their natural living environment can a more accurate assessment of health be obtained. The ability to assess hearing health, social engagement, cognitive, and physical health in natural living environments is becoming possible with the advent of ubiquitous sensing capabilities.

PURPOSE

Here we discuss some of the work that has been done by our group and others that may be of use to the field of audiology e-health. The purpose of this article is not to present new experimental data, but rather to describe a new method of using advanced in-home sensing techniques to better understand how hearing diagnostics, interventions, and rehabilitation influence the lives and behaviors of patients.