Artificial Intelligence in Decision Support Systems for Type 1 Diabetes

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.

Application of machine learning in affordable and accessible insulin management for type 1 and 2 diabetes: A comprehensive review

Proper insulin management is vital for maintaining stable blood sugar levels and preventing complications associated with diabetes. However, the soaring costs of insulin present significant challenges to ensuring affordable management. This paper conducts a comprehensive review of current literature on the application of machine learning (ML) in insulin management for diabetes patients, particularly focusing on enhancing affordability and accessibility within the United States. The review encompasses various facets of insulin management, including dosage calculation and response, prediction of blood glucose and insulin sensitivity, initial insulin estimation, resistance prediction, treatment adherence, complications, hypoglycemia prediction, and lifestyle modifications. Additionally, the study identifies key limitations in the utilization of ML within the insulin management literature and suggests future research directions aimed at furthering accessible and affordable insulin treatments. These proposed directions include exploring insurance coverage, optimizing insulin type selection, assessing the impact of biosimilar insulin and market competition, considering mental health factors, evaluating insulin delivery options, addressing cost-related issues affecting insulin usage and adherence, and selecting appropriate patient cost-sharing programs. By examining the potential of ML in addressing insulin management affordability and accessibility, this work aims to envision improved and cost-effective insulin management practices. It not only highlights existing research gaps but also offers insights into future directions, guiding the development of innovative solutions that have the potential to revolutionize insulin management and benefit patients reliant on this life-saving treatment.

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.

Community pharmacists awareness, perceptions, and opinions of artificial intelligence: A cross-sectional study in Riyadh, Saudi Arabia

BACKGROUND

Several revolutions are currently taking place in the healthcare industry to provide accurate, reliable, and valid healthcare to patients. Among these is artificial intelligence (AI).

OBJECTIVE

This study aimed to assess the CP's awareness, perceptions, and opinions of AI in health care among community pharmacists.

METHODS

This cross-sectional survey-based study was conducted over 3 months in 2023 using structured prevalidated 34 items questionnaires.

RESULTS

In this study, 94.5% (n= 258) of the CPs were aware of AI, yet 25.6% (n= 70) believed that AI would eventually replace healthcare professionals. However, 63.4% (n= 173) of the CPs concurred that AI is a technology that supports healthcare workers. 12.8% of the CPs believed that there is a risk of losing their jobs if AI is widely used in Saudi Arabia, but 68.9% (n= 188) of them considered that healthcare professionals will benefit from the extensive use of AI. Eighty-four percent of CPs (n= 232) agreed or strongly agreed that AI decreases drug mistakes in clinical practice. Similarly, 86% of the CPs (n= 235) concurred that AI makes it easier for patients to access the service. In contrast, almost 58% of the CPs (n= 232) agreed that AI makes it easier for healthcare professionals to acquire information, and 87.9% of the CPs (n= 240) said that AI helps them make better decisions.

CONCLUSION

This study concluded that most of the CPs were aware of AI and agreed that AI is a tool that helps healthcare professionals. In addition, the majority of the CPs thought that AI adoption in healthcare practice will benefit healthcare practitioners.

Applications of Clinical Decision Support Systems in Diabetes Care: Scoping Review

BACKGROUND

Providing comprehensive and individualized diabetes care remains a significant challenge in the face of the increasing complexity of diabetes management and a lack of specialized endocrinologists to support diabetes care. Clinical decision support systems (CDSSs) are progressively being used to improve diabetes care, while many health care providers lack awareness and knowledge about CDSSs in diabetes care. A comprehensive analysis of the applications of CDSSs in diabetes care is still lacking.

OBJECTIVE

This review aimed to summarize the research landscape, clinical applications, and impact on both patients and physicians of CDSSs in diabetes care.

METHODS

We conducted a scoping review following the Arksey and O'Malley framework. A search was conducted in 7 electronic databases to identify the clinical applications of CDSSs in diabetes care up to June 30, 2022. Additional searches were conducted for conference abstracts from the period of 2021-2022. Two researchers independently performed the screening and data charting processes.

RESULTS

Of 11,569 retrieved studies, 85 (0.7%) were included for analysis. Research interest is growing in this field, with 45 (53%) of the 85 studies published in the past 5 years. Among the 58 (68%) out of 85 studies disclosing the underlying decision-making mechanism, most CDSSs (44/58, 76%) were knowledge based, while the number of non-knowledge-based systems has been increasing in recent years. Among the 81 (95%) out of 85 studies disclosing application scenarios, the majority of CDSSs were used for treatment recommendation (63/81, 78%). Among the 39 (46%) out of 85 studies disclosing physician user types, primary care physicians (20/39, 51%) were the most common, followed by endocrinologists (15/39, 39%) and nonendocrinology specialists (8/39, 21%). CDSSs significantly improved patients' blood glucose, blood pressure, and lipid profiles in 71% (45/63), 67% (12/18), and 38% (8/21) of the studies, respectively, with no increase in the risk of hypoglycemia.

CONCLUSIONS

CDSSs are both effective and safe in improving diabetes care, implying that they could be a potentially reliable assistant in diabetes care, especially for physicians with limited experience and patients with limited access to medical resources.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

RR2-10.37766/inplasy2022.9.0061.

Transform diabetes care with precision medicine

Background and Aims

Diabetes is a global concern. This article took a closer look at diabetes and precision medicine.

Methods

A literature search of studies related to the use of precision medicine in diabetes care was conducted in various databases (PubMed, Google Scholar, and Scopus).

Results

Precision medicine encompasses the integration of a wide array of personal data, including clinical, lifestyle, genetic, and various biomarker information. Its goal is to facilitate tailored treatment approaches using contemporary diagnostic and therapeutic techniques that specifically target patients based on their genetic makeup, molecular markers, phenotypic traits, or psychosocial characteristics. This article not only highlights significant advancements but also addresses key challenges, particularly focusing on the technologies that contribute to the realization of personalized and precise diabetes care.

Conclusion

For the successful implementation of precision diabetes medicine, collaboration and coordination among multiple stakeholders are crucial.

The importance of interpreting machine learning models for blood glucose prediction in diabetes: an analysis using SHAP

Machine learning has become a popular tool for learning models of complex dynamics from biomedical data. In Type 1 Diabetes (T1D) management, these models are increasingly been integrated in decision support systems (DSS) to forecast glucose levels and provide preventive therapeutic suggestions, like corrective insulin boluses (CIB), accordingly. Typically, models are chosen based on their prediction accuracy. However, since patient safety is a concern in this application, the algorithm should also be physiologically sound and its outcome should be explainable. This paper aims to discuss the importance of using tools to interpret the output of black-box models in T1D management by presenting a case-of-study on the selection of the best prediction algorithm to integrate in a DSS for CIB suggestion. By retrospectively "replaying" real patient data, we show that two long-short term memory neural networks (LSTM) (named p-LSTM and np-LSTM) with similar prediction accuracy could lead to different therapeutic decisions. An analysis with SHAP-a tool for explaining black-box models' output-unambiguously shows that only p-LSTM learnt the physiological relationship between inputs and glucose prediction, and should therefore be preferred. This is verified by showing that, when embedded in the DSS, only p-LSTM can improve patients' glycemic control.

A Markov Model of Gap Occurrence in Continuous Glucose Monitoring Data for Realistic in Silico Clinical Trials

BACKGROUND AND OBJECTIVE

Continuous glucose monitoring (CGM) sensors measure interstitial glucose concentration every 1-5 min for days or weeks. New CGM-based diabetes therapies are often tested in in silico clinical trials (ISCTs) using diabetes simulators. Accurate models of CGM sensor inaccuracies and failures could help improve the realism of ISCTs. However, the modeling of CGM failures has not yet been fully addressed in the literature. This work aims to develop a mathematical model of CGM gaps, i.e., occasional portions of missing data generated by temporary sensor errors (e.g., excessive noise or artifacts).

METHODS

Two datasets containing CGM traces collected in 167 adults and 205 children, respectively, using the Dexcom G6 sensor (Dexcom Inc., San Diego, CA) were used. Four Markov models, of increasing complexity, were designed to describe three main characteristics: number of gaps for each sensor, gap distribution in the monitoring days, and gap duration. Each model was identified on a portion of each dataset (training set). The remaining portion of each dataset (real test set) was used to evaluate model performance through a Monte Carlo simulation approach. Each model was used to generate 100 simulated test sets with the same size as the real test set. The distributions of gap characteristics on the simulated test sets were compared with those observed on the real test set, using the two-sample Kolmogorov-Smirnov test and the Jensen-Shannon divergence.

RESULTS

A six-state Markov model, having two states to describe normal sensor operation and four states to describe gap occurrence, achieved the best results. For this model, the Kolmogorov-Smirnov test found no significant differences between the distribution of simulated and real gap characteristics. Moreover, this model obtained significantly lower Jensen-Shannon divergence values than the other models.

CONCLUSIONS

A Markov model describing CGM gaps was developed and validated on two real datasets. The model describes well the number of gaps for each sensor, the gap distribution over monitoring days, and the gap durations. Such a model can be integrated into existing diabetes simulators to realistically simulate CGM gaps in ISCTs and thus enable the development of more effective and robust diabetes management strategies.

Artificial intelligence for home monitoring devices

PURPOSE OF REVIEW

Home monitoring in ophthalmology is appropriate for disease stages requiring frequent monitoring or rapid intervention, for example, neovascular age-related macular degeneration (AMD) and glaucoma, where the balance between frequent hospital attendance versus risk of late detection is a constant challenge. Artificial intelligence approaches are well suited to address some challenges of home monitoring.

RECENT FINDINGS

Ophthalmic data collected at home have included functional (e.g. perimetry), biometric (e.g. intraocular pressure), and imaging [e.g. optical coherence tomography (OCT)] data. Potential advantages include early detection/intervention, convenience, cost, and visual outcomes. Artificial intelligence can assist with home monitoring workflows by handling large data volumes from frequent testing, compensating for test quality, and extracting useful metrics from complex data. Important use cases include machine learning applied to hyperacuity self-testing for detecting neovascular AMD and deep learning applied to OCT data for quantifying retinal fluid.

SUMMARY

Home monitoring of health conditions is useful for chronic diseases requiring rapid intervention or frequent data sampling to decrease risk of irreversible vision loss. Artificial intelligence may facilitate accurate, frequent, large-scale home monitoring, if algorithms are integrated safely into workflows. Clinical trials and economic evaluations are important to demonstrate the value of artificial intelligence-based home monitoring, towards improved visual outcomes.

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.

DiaTrend: A dataset from advanced diabetes technology to enable development of novel analytic solutions

Objective digital data is scarce yet needed in many domains to enable research that can transform the standard of healthcare. While data from consumer-grade wearables and smartphones is more accessible, there is critical need for similar data from clinical-grade devices used by patients with a diagnosed condition. The prevalence of wearable medical devices in the diabetes domain sets the stage for unique research and development within this field and beyond. However, the scarcity of open-source datasets presents a major barrier to progress. To facilitate broader research on diabetes-relevant problems and accelerate development of robust computational solutions, we provide the DiaTrend dataset. The DiaTrend dataset is composed of intensive longitudinal data from wearable medical devices, including a total of 27,561 days of continuous glucose monitor data and 8,220 days of insulin pump data from 54 patients with diabetes. This dataset is useful for developing novel analytic solutions that can reduce the disease burden for people living with diabetes and increase knowledge on chronic condition management in outpatient settings.

Using Constrained-Disorder Principle-Based Systems to Improve the Performance of Digital Twins in Biological Systems

Digital twins are computer programs that use real-world data to create simulations that predict the performance of processes, products, and systems. Digital twins may integrate artificial intelligence to improve their outputs. Models for dealing with uncertainties and noise are used to improve the accuracy of digital twins. Most currently used systems aim to reduce noise to improve their outputs. Nevertheless, biological systems are characterized by inherent variability, which is necessary for their proper function. The constrained-disorder principle defines living systems as having a disorder as part of their existence and proper operation while kept within dynamic boundaries. In the present paper, we review the role of noise in complex systems and its use in bioengineering. We describe the use of digital twins for medical applications and current methods for dealing with noise and uncertainties in modeling. The paper presents methods to improve the accuracy and effectiveness of digital twin systems by continuously implementing variability signatures while simultaneously reducing unwanted noise in their inputs and outputs. Accounting for the noisy internal and external environments of complex biological systems is necessary for the future design of improved, more accurate digital twins.

Current Technologies for Managing Type 1 Diabetes Mellitus and Their Impact on Quality of Life-A Narrative Review

Type 1 diabetes mellitus is a chronic autoimmune disease that affects millions of people and generates high healthcare costs due to frequent complications when inappropriately managed. Our paper aimed to review the latest technologies used in T1DM management for better glycemic control and their impact on daily life for people with diabetes. Continuous glucose monitoring systems provide a better understanding of daily glycemic variations for children and adults and can be easily used. These systems diminish diabetes distress and improve diabetes control by decreasing hypoglycemia. Continuous subcutaneous insulin infusions have proven their benefits in selected patients. There is a tendency to use more complex systems, such as hybrid closed-loop systems that can modulate insulin infusion based on glycemic readings and artificial intelligence-based algorithms. It can help people manage the burdens associated with T1DM management, such as fear of hypoglycemia, exercising, and long-term complications. The future is promising and aims to develop more complex ways of automated control of glycemic levels to diminish the distress of individuals living with diabetes.

Clinical Experience of Using Telemedicine for the Management of Patients Using Continuous Subcutaneous Insulin Infusion in a Highly Complex Latin American Hospital

INTRODUCTION

Continuous subcutaneous insulin infusion (CSII) has emerged as a potential solution for diabetes management during the pandemic, as it reduces the need for in-person visits and allows for remote monitoring of patients. Telemedicine has also become increasingly important in the management of diabetes during the pandemic, as it allows healthcare providers to provide remote consultations and support. Here, we discuss the implications of this approach for diabetes management beyond the pandemic, including the potential for increased access to care and improved patient outcomes.

METHODS

We performed a longitudinal observational study between 1 March 2020 and 31 December 2020 to evaluate glycemic parameters in diabetic patients with CSII in a telehealth service. Glycemic parameters were time in range (TIR), time above range, time below range, mean daily glucose, glucose management indicator (GMI), and glycemic variability control.

RESULTS

A total of 36 patients were included in the study, with 29 having type 1 diabetes and 6 having type 2 diabetes. The study found that the proportion of patients achieving target glucose variability and GMI remained unchanged during follow-up. However, in patients with type 2 diabetes, the time in target range increased from 70% to 80%, and the time in hyperglycemia decreased from 2% to 0%.

CONCLUSIONS

The results of this study suggest that telemedicine is a strategy for maintaining glycemic control in patients using CSII. However, the lack of access to the internet and adequate telemonitoring devices make it difficult to use on a large scale in emerging countries like ours.

Safety and Efficacy of an Adaptive Bolus Calculator for Type 1 Diabetes: a Randomised Control Cross over Study

Background The Advanced Bolus Calculator for Type 1 Diabetes (ABC4D) is a decision support system employing the artificial intelligence technique of case-based reasoning to adapt and personalise insulin bolus doses. The integrated system comprises a smartphone application and clinical web portal. We aimed to assess safety and efficacy of the ABC4D (intervention) compared to a non-adaptive bolus calculator (control). Methods This was a prospective randomised controlled crossover study. Following a 2-week run-in period, participants were randomised to ABC4D or control for 12 weeks. After a 6-week washout period, participants crossed over for 12 weeks. The primary outcome was difference in percentage (%) time in range (TIR) (3.9-10.0 mmol/L (70-180mg/dL)) change during the daytime (07:00-22:00) between groups. Results 37 adults with type 1 diabetes on multiple daily injections of insulin were randomised, median (IQR) age 44.7 (28.2-55.2) years, diabetes duration 15.0 (9.5-29.0) years, HbA1C 61.0 (58.0-67.0) mmol/mol (7.7 (7.5-8.3)%). Data from 33 participants were analysed. There was no significant difference in daytime %TIR change with ABC4D compared to control (median (IQR) +0.1 (-2.6 to + 4.0)% versus +1.9 (-3.8 to + 10.1)%; p = 0.53). Participants accepted fewer meal dose recommendations in the intervention compared to control (78.7 (55.8-97.6)% versus 93.5 (73.8-100)%; p = 0.009) with a greater reduction in insulin dosage from that recommended. Conclusion The ABC4D is safe for adapting insulin bolus doses and provided the same level of glycaemic control as the non-adaptive bolus calculator. Results suggest that participants did not follow ABC4D recommendations as frequently as control, impacting its effectiveness.

A Federated Learning-Inspired Evolutionary Algorithm: Application to Glucose Prediction

In this paper, we propose an innovative Federated Learning-inspired evolutionary framework. Its main novelty is that this is the first time that an Evolutionary Algorithm is employed on its own to directly perform Federated Learning activity. A further novelty resides in the fact that, differently from the other Federated Learning frameworks in the literature, ours can efficiently deal at the same time with two relevant issues in Machine Learning, i.e., data privacy and interpretability of the solutions. Our framework consists of a master/slave approach in which each slave contains local data, protecting sensible private data, and exploits an evolutionary algorithm to generate prediction models. The master shares through the slaves the locally learned models that emerge on each slave. Sharing these local models results in global models. Being that data privacy and interpretability are very significant in the medical domain, the algorithm is tested to forecast future glucose values for diabetic patients by exploiting a Grammatical Evolution algorithm. The effectiveness of this knowledge-sharing process is assessed experimentally by comparing the proposed framework with another where no exchange of local models occurs. The results show that the performance of the proposed approach is better and demonstrate the validity of its sharing process for the emergence of local models for personal diabetes management, usable as efficient global models. When further subjects not involved in the learning process are considered, the models discovered by our framework show higher generalization capability than those achieved without knowledge sharing: the improvement provided by knowledge sharing is equal to about 3.03% for precision, 1.56% for recall, 3.17% for F1, and 1.56% for accuracy. Moreover, statistical analysis reveals the statistical superiority of model exchange with respect to the case of no exchange taking place.

Evaluation of a Web-Based Simulation Tool for Self-Management Support in Type 1 Diabetes: A Pilot Study

OBJECTIVE

To develop and evaluate a novel Web-based Simulation Tool (WST) that brings simulation technologies to people with Type 1 Diabetes (T1D), enabling unique patient-data interactions seamlessly on a daily basis.

METHODS

A pilot clinical trial was conducted to assess system usability. The study consisted of one week of observation (Phase 1) and four weeks of interaction with WST (Phase 2). Responses to Technology Acceptance (TA) and Diabetes Distress Scale (DDS) questionnaires were collected, and follow-up interviews were conducted after Phase 2.

RESULTS

Fifteen participants with T1D using Control-IQ technology (age: 36 ± 13 years, HbA1c: 6.5% ± 0.7%) completed all study procedures. Generated simulation models achieved a median Mean Absolute Relative Difference (MARD) of 6.8% [interquartile range, IQR: 5.1%, 9.1%]. A decrease in expected benefits (likely explained by issues with the third-party data collection system) and an increase in expected burdens were observed. On a 1-5 scale, ease of use, trust, and usefulness scores were 3 [3,4], 4 [3,4], and 4 [3,4], respectively. Time below 70 mg/dL decreased between Phases 1 and 2 (1.6% [0.7%,3.7%] vs 0.8% [0.5%,3.0%]). A reduction in mean emotional burden was also observed (2.5 ± 1.1 vs 2.1 ± 0.8).

CONCLUSIONS

Results indicate that there was a learning curve to WST, but also that most participants trusted the system and found it useful in their diabetes care.

SIGNIFICANCE

Simulation technologies like WST could be used by educators and patients to facilitate diabetes self-management, leading to better diabetes literacy and reducing associated distress.

Continuous glucose monitoring for children with hypoglycaemia: Evidence in 2023

In 2023, childhood hypoglycaemia remains a major public health problem and significant risk factor for consequent adverse neurodevelopment. Irrespective of the underlying cause, key elements of clinical management include the detection, prediction and prevention of episodes of hypoglycaemia. These tasks are increasingly served by Continuous Glucose Monitoring (CGM) devices that measure subcutaneous glucose at near-continuous frequency. While the use of CGM in type 1 diabetes is well established, the evidence for widespread use in rare hypoglycaemia disorders is less than convincing. However, in the few years since our last review there have been multiple developments and increased user feedback, requiring a review of clinical application. Despite advances in device technology, point accuracy of CGM remains low for children with non-diabetes hypoglycaemia. Simple provision of CGM devices has not replicated the efficacy seen in those with diabetes and is yet to show benefit. Machine learning techniques for hypoglycaemia prevention have so far failed to demonstrate sufficient prediction accuracy for real world use even in those with diabetes. Furthermore, access to CGM globally is restricted by costs kept high by the commercially-driven speed of technical innovation. Nonetheless, the ability of CGM to digitally phenotype disease groups has led to a better understanding of natural history of disease, facilitated diagnoses and informed changes in clinical management. Large CGM datasets have prompted re-evaluation of hypoglycaemia incidence and facilitated improved trial design. Importantly, an individualised approach and focus on the behavioural determinants of hypoglycaemia has led to real world reduction in hypoglycaemia. In this state of the art review, we critically analyse the updated evidence for use of CGM in non-diabetic childhood hypoglycaemia disorders since 2020 and provide suggestions for qualified use.

The behaviour change behind a successful pilot of hypoglycaemia reduction with HYPO-CHEAT

Background

Children with hypoglycaemia disorders, such as congenital hyperinsulinism (CHI), are at constant risk of hypoglycaemia (low blood sugars) with the attendant risk of brain injury. Current approaches to hypoglycaemia detection and prevention vary from fingerprick glucose testing to the provision of continuous glucose monitoring (CGM) to machine learning (ML) driven glucose forecasting. Recent trends for ML have had limited success in preventing free-living hypoglycaemia, due to a focus on increasingly accurate glucose forecasts and a failure to acknowledge the human in the loop and the essential step of changing behaviour. The wealth of evidence from the fields of behaviour change and persuasive technology (PT) allows for the creation of a theory-informed and technologically considered approach.

Objectives

We aimed to create a PT that would overcome the identified barriers to hypoglycaemia prevention for those with CHI to focus on proactive prevention rather than commonly used reactive approaches.

Methods

We used the behaviour change technique taxonomy and persuasive systems design models to create HYPO-CHEAT (HYpoglycaemia-Prevention-thrOugh-Cgm-HEatmap-Assisted-Technology): a novel approach that presents aggregated CGM data in simple visualisations. The resultant ease of data interpretation is intended to facilitate behaviour change and subsequently reduce hypoglycaemia.

Results

HYPO-CHEAT was piloted in 10 patients with CHI over 12 weeks and successfully identified weekly patterns of hypoglycaemia. These patterns consistently correlated with identifiable behaviours and were translated into both a change in proximal fingerprick behaviour and ultimately, a significant reduction in aggregated hypoglycaemia from 7.1% to 5.4% with four out of five patients showing clinically meaningful reductions in hypoglycaemia.

Conclusions

We have provided pilot data of a new approach to hypoglycaemia prevention that focuses on proactive prevention and behaviour change. This approach is personalised for individual patients with CHI and is a first step in changing our approach to hypoglycaemia prevention in this group.

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.

Assessment of Seasonal Stochastic Local Models for Glucose Prediction without Meal Size Information under Free-Living Conditions

Accurate blood glucose (BG) forecasting is key in diabetes management, as it allows preventive actions to mitigate harmful hypoglycemic/hyperglycemic episodes. Considering the encouraging results obtained by seasonal stochastic models in proof-of-concept studies, this work assesses the methodology in two datasets (open-loop and closed-loop) recorded in free-living conditions. First, similar postprandial glycemic profiles are grouped together with fuzzy C-means clustering. Then, a seasonal stochastic model is identified for each cluster. Finally, real-time BG forecasting is performed by weighting each model’s prediction. The proposed methodology (named C-SARIMA) is compared to other linear and nonlinear black-box methods: autoregressive integrated moving average (ARIMA), its variant with input (ARIMAX), a feed-forward neural network (NN), and its modified version (NN-X) fed by BG, insulin, and carbohydrates (timing and dosing) information for several prediction horizons (PHs). In the open-loop dataset, C-SARIMA grants a median root-mean-squared error (RMSE) of 20.13 mg/dL (PH = 30) and 27.23 mg/dL (PH = 45), not significantly different from ARIMA and NN. Over a longer PH, C-SARIMA achieves an RMSE = 31.96 mg/dL (PH = 60) and RMSE = 33.91 mg/dL (PH = 75), significantly outperforming the ARIMA and NN, without significant differences from the ARIMAX for PH ≥ 45 and the NN-X for PH ≥ 60. Similar results hold on the closed-loop dataset: for PH = 30 and 45 min, the C-SARIMA achieves an RMSE = 21.63 mg/dL and RMSE = 29.67 mg/dL, not significantly different from the ARIMA and NN. On longer PH, the C-SARIMA outperforms the ARIMA for PH > 45 and the NN for PH > 60 without significant differences from the ARIMAX for PH ≥ 45. Although using less input information, the C-SARIMA achieves similar performance to other prediction methods such as the ARIMAX and NN-X and outperforming the CGM-only approaches on PH > 45min.

HYPO-CHEAT's aggregated weekly visualisations of risk reduce real world hypoglycaemia

Background

Children with congenital hyperinsulinism (CHI) are at constant risk of hypoglycaemia with the attendant risk of brain injury. Current hypoglycaemia prevention methods centre on the prediction of a continuous glucose variable using machine learning (ML) processing of continuous glucose monitoring (CGM). This approach ignores repetitive and predictable behavioural factors and is dependent upon ongoing CGM. Thus, there has been very limited success in reducing real-world hypoglycaemia with a ML approach in any condition.

Objectives

We describe the development of HYPO-CHEAT (HYpoglycaemia-Prevention-thrOugh-CGM-HEatmap-Technology), which is designed to overcome these limitations by describing weekly hypoglycaemia risk. We tested HYPO-CHEAT in a real-world setting to evaluate change in hypoglycaemia.

Methods

HYPO-CHEAT aggregates individual CGM data to identify weekly hypoglycaemia patterns. These are visualised via a hypoglycaemia heatmap along with actionable interpretations and targets. The algorithm is iterative and reacts to anticipated changing patterns of hypoglycaemia. HYPO-CHEAT was compared with Dexcom Clarity's pattern identification and Facebook Prophet's forecasting algorithm using data from 10 children with CHI using CGM for 12 weeks. HYPO-CHEAT's efficacy was assessed via change in time below range (TBR).

Results

HYPO-CHEAT identified hypoglycaemia patterns in all patients. Dexcom Clarity identified no patterns. Predictions from Facebook Prophet were inconsistent and difficult to interpret. Importantly, the patterns identified by HYPO-CHEAT matched the lived experience of all patients, generating new and actionable understanding of the cause of hypos. This facilitated patients to significantly reduce their time in hypoglycaemia from 7.1% to 5.4% even when real-time CGM data was removed.

Conclusions

HYPO-CHEAT's personalised hypoglycaemia heatmaps reduced total and targeted TBR even when CGM was reblinded. HYPO-CHEAT offers a highly effective and immediately available personalised approach to prevent hypoglycaemia and empower patients to self-care.

Application of Artificial Intelligence in Discovery and Development of Anticancer and Antidiabetic Therapeutic Agents

Spectacular developments in molecular and cellular biology have led to important discoveries in cancer research. Despite cancer is one of the major causes of morbidity and mortality globally, diabetes is one of the most leading sources of group of disorders. Artificial intelligence (AI) has been considered the fourth industrial revolution machine. The most major hurdles in drug discovery and development are the time and expenditures required to sustain the drug research pipeline. Large amounts of data can be explored and generated by AI, which can then be converted into useful knowledge. Because of this, the world's largest drug companies have already begun to use AI in their drug development research. In the present era, AI has a huge amount of potential for the rapid discovery and development of new anticancer drugs. Clinical studies, electronic medical records, high-resolution medical imaging, and genomic assessments are just a few of the tools that could aid drug development. Large data sets are available to researchers in the pharmaceutical and medical fields, which can be analyzed by advanced AI systems. This review looked at how computational biology and AI technologies may be utilized in cancer precision drug development by combining knowledge of cancer medicines, drug resistance, and structural biology. This review also highlighted a realistic assessment of the potential for AI in understanding and managing diabetes.

A Brief Review on the Evolution of Technology in Exercise and Sport in Type 1 Diabetes: Past, Present, and Future

One hundred years ago, insulin was first used to successfully lower blood glucose levels in young people living with what was then called juvenile diabetes. While insulin was not a cure for diabetes, it allowed individuals to resume a near normal life and have some freedom to eat more liberally and gain the strength they needed to live a more active lifestyle. Since then, a number of therapeutic and technical advances have arisen to further improve the health and wellbeing of individuals living with type 1 diabetes, allowing many to participate in sport at the local, regional, national or international level of competition. This review and commentary highlights some of the key advances in diabetes management in sport over the last 100 years since the discovery of insulin.

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.

A personalized blood glucose level prediction model with a fine-tuning strategy: A proof-of-concept study

BACKGROUND

The accurate prediction of blood glucose (BG) level is still a challenge for diabetes management. This is due to various factors such as diet, personal physiological characteristics, stress, and activities influence changes in BG level. To develop an accurate BG level predictive model, we propose a personalized model based on a convolutional neural network (CNN) with a fine-tuning strategy.

METHODS

We utilized continuous glucose monitoring (CGM) datasets from 1052 professional CGM sessions and split them into three groups according to type 1, type 2, and gestational diabetes mellitus (T1DM, T2DM, and GDM, respectively). During the preprocessing, only CGM data points were utilized, and future BG levels of four different prediction horizons (PHs, 15, 30, 45, and 60 min) were used as output. In training, we trained a general CNN and a multi-output random forest regressor using a hold-out method for each group. Next, we developed two personalized models: (1) by fine-tuning the general CNN on partial sample points of each CGM dataset, and (2) by learning a CNN from scratch on the points.

RESULTS

For all groups, the fine-tuned CNN showed the lowest average root mean squared error, average mean absolute percentage error, highest average time gain (PH = 15 and 60 min in T1DM) and highest percentage in region A of Clarke error grid analysis at all PHs. In the performance comparison between the fine-tuned CNN and other models, we found that the fine-tuned CNN improved the performance of the general CNN in most cases and outperformed the scratch CNN at all PHs in all groups, making the fine-tuning strategy was useful for accurate BG level prediction. We analyzed all cases of four predictive patterns in each group, and found that the input BG level trend and the BG level at the time of prediction were related to the future BG level trend.

CONCLUSIONS

We demonstrated the efficacy of the fine-tuning method in a large number of CGM datasets and analyzed the four predictive patterns. Therefore, we believe that the proposed method will significantly contribute to the development of an accurate personalized model and the analysis for its predictions.

Artificial intelligence and diabetes technology: A review

Artificial intelligence (AI) is widely discussed in the popular literature and is portrayed as impacting many aspects of human life, both in and out of the workplace. The potential for revolutionizing healthcare is significant because of the availability of increasingly powerful computational platforms and methods, along with increasingly informative sources of patient data, both in and out of clinical settings. This review aims to provide a realistic assessment of the potential for AI in understanding and managing diabetes, accounting for the state of the art in the methodology and medical devices that collect data, process data, and act accordingly. Acknowledging that many conflicting definitions of AI have been put forth, this article attempts to characterize the main elements of the field as they relate to diabetes, identifying the main perspectives and methods that can (i) affect basic understanding of the disease, (ii) affect understanding of risk factors (genetic, clinical, and behavioral) of diabetes development, (iii) improve diagnosis, (iv) improve understanding of the arc of disease (progression and personal/societal impact), and finally (v) improve treatment.

How can artificial intelligence be used for peptidomics?

INTRODUCTION

Peptidomics is an emerging field of omics sciences using advanced isolation, analysis, and computational techniques that enable qualitative and quantitative analyses of various peptides in biological samples. Peptides can act as useful biomarkers and as therapeutic molecules for diseases.

AREAS COVERED

The use of therapeutic peptides can be predicted quickly and efficiently using data-driven computational methods, particularly artificial intelligence (AI) approach. Various AI approaches are useful for peptide-based drug discovery, such as support vector machine, random forest, extremely randomized trees, and other more recently developed deep learning methods. AI methods are relatively new to the development of peptide-based therapies, but these techniques already become essential tools in protein science by dissecting novel therapeutic peptides and their functions (Figure 1).[Figure: see text].

EXPERT OPINION

Researchers have shown that AI models can facilitate the development of peptidomics and selective peptide therapies in the field of peptide science. Biopeptide prediction is important for the discovery and development of successful peptide-based drugs. Due to their ability to predict therapeutic roles based on sequence details, many AI-dependent prediction tools have been developed (Figure 1).

Comparative effectiveness of team-based care with a clinical decision support system versus team-based care alone on cardiovascular risk reduction among patients with diabetes: Rationale and design of the D4C trial

BACKGROUND

Diabetes has become a major public health challenge worldwide, especially in low- and middle-income countries (LMICs). Uncontrolled hyperglycemia, hypertension, and dyslipidemia major risk factors for all-cause mortality and cardiovascular disease (CVD) are common in patients with diabetes in China. We propose to compare the effectiveness of team-based care plus a clinical decision support system (CDSS) with team-based care alone on glycemic, blood pressure (BP), and lipid control, and clinical CVD reduction among patients with type-2 diabetes and at high risk for CVD.

METHODS

The Diabetes Complication Control in Community Clinics (D4C) study is a cluster-randomized trial conducted among 38 community health centers in Xiamen City, China. Nineteen clinics have been randomly assigned to team-based care plus CDSS and 19 to team-based care alone. Team-based care includes primary care providers, health coaches, and diabetes specialists working collaboratively with patients to achieve shared treatment goals for CVD risk factor reduction. The CDSS integrates guideline-based treatment algorithms for glycemic, BP, and lipid control, along with a patient's medical history and insurance policy, to recommend treatment and follow-up plans. In phase 1, the co-primary outcomes are mean reduction in glycated hemoglobin (HbA1c), systolic BP (SBP), and low-density lipoprotein (LDL)-cholesterol over 18 months, and the proportion of patients with controlled HbA1c, SBP, and LDL-cholesterol at 18 months' between the 2 comparison groups. In phase 2, the primary outcome is the difference in major CVD incidence (non-fatal stroke, non-fatal myocardial infarction, hospitalized heart failure, and CVD mortality) between the 2 comparison groups. Mean reduction in HbA1c, SBP, and LDL-cholesterol levels will be simultaneously modeled for a single overall treatment effect.

CONCLUSION

The D4C trial will generate evidence on whether a CDSS will further reduce the CVD burden among patients with diabetes beyond team-based care at community clinics. If proven effective, this implementation strategy could be scaled up within primary care settings in China and other LMICs to reduce CVD incidence and mortality among patients with diabetes.

A Multitask Learning Approach to Personalised Blood Glucose Prediction

Blood glucose prediction algorithms are key tools in the development of decision support systems and closed-loop insulin delivery systems for blood glucose control in diabetes. Deep learning models have provided leading results among machine learning algorithms to date in glucose prediction. However these models typically require large amounts of data to obtain best personalised glucose prediction results. Multitask learning facilitates an approach for leveraging data from multiple subjects while still learning accurate personalised models. In this work we present results comparing the effectiveness of multitask learning over sequential transfer learning, and learning only on subject-specific data with neural networks and support vector regression. The multitask learning approach shows consistent leading performance in predictive metrics at both short-term and long-term prediction horizons. We obtain a predictive accuracy (RMSE) of 18.8 2.3, 25.3 2.9, 31.8 3.9, 41.2 4.5, 47.2 4.6 mg/dL at 30, 45, 60, 90, and 120 min prediction horizons respectively, with at least 93\% clinically acceptable predictions using the Clarke Error Grid (EGA) at each prediction horizon. We also identify relevant prior information such as glycaemic variability that can be incorporated to improve predictive performance at long-term prediction horizons. Furthermore, we demonstrate consistent performance - 5% change in both RMSE and EGA (Zone A) - in rare cases of adverse glycaemic events with 1-6 weeks of training data. In conclusion, a multitask approach can allow for deploying personalised models even with significantly less subject-specific data without compromising performance.

Probabilistic Model of Transition between Categories of Glucose Profiles in Patients with Type 1 Diabetes Using a Compositional Data Analysis Approach

The time spent in glucose ranges is a common metric in type 1 diabetes (T1D). As the time in one day is finite and limited, Compositional Data (CoDa) analysis is appropriate to deal with times spent in different glucose ranges in one day. This work proposes a CoDa approach applied to glucose profiles obtained from six T1D patients using continuous glucose monitor (CGM). Glucose profiles of 24-h and 6-h duration were categorized according to the relative interpretation of time spent in different glucose ranges, with the objective of presenting a probabilistic model of prediction of category of the next 6-h period based on the category of the previous 24-h period. A discriminant model for determining the category of the 24-h periods was obtained, achieving an average above 94% of correct classification. A probabilistic model of transition between the category of the past 24-h of glucose to the category of the future 6-h period was obtained. Results show that the approach based on CoDa is suitable for the categorization of glucose profiles giving rise to a new analysis tool. This tool could be very helpful for patients, to anticipate the occurrence of potential adverse events or undesirable variability and for physicians to assess patients' outcomes and then tailor their therapies.

Safety and Feasibility of the PEPPER Adaptive Bolus Advisor and Safety System: A Randomized Control Study

Background: The Patient Empowerment through Predictive Personalized Decision Support (PEPPER) system provides personalized bolus advice for people with type 1 diabetes. The system incorporates an adaptive insulin recommender system (based on case-based reasoning, an artificial intelligence methodology), coupled with a safety system, which includes predictive glucose alerts and alarms, predictive low-glucose suspend, personalized carbohydrate recommendations, and dynamic bolus insulin constraint. We evaluated the safety and efficacy of the PEPPER system compared to a standard bolus calculator. Methods: This was an open-labeled multicenter randomized controlled crossover study. Following 4-week run-in, participants were randomized to PEPPER/Control or Control/PEPPER in a 1:1 ratio for 12 weeks. Participants then crossed over after a washout period. The primary end-point was percentage time in range (TIR, 3.9-10.0 mmol/L [70-180 mg/dL]). Secondary outcomes included glycemic variability, quality of life, and outcomes on the safety system and insulin recommender. Results: Fifty-four participants on multiple daily injections (MDI) or insulin pump completed the run-in period, making up the intention-to-treat analysis. Median (interquartile range) age was 41.5 (32.3-49.8) years, diabetes duration 21.0 (11.5-26.0) years, and HbA1c 61.0 (58.0-66.1) mmol/mol. No significant difference was observed for percentage TIR between the PEPPER and Control groups (62.5 [52.1-67.8] % vs. 58.4 [49.6-64.3] %, respectively, P = 0.27). For quality of life, participants reported higher perceived hypoglycemia with the PEPPER system despite no objective difference in time spent in hypoglycemia. Conclusions: The PEPPER system was safe, but did not change glycemic outcomes, compared to control. There is wide scope for integrating PEPPER into routine diabetes management for pump and MDI users. Further studies are required to confirm overall effectiveness. Clinical trial registration: ClinicalTrials.gov NCT03849755.

The Need for Ethnoracial Equity in Artificial Intelligence for Diabetes Management: Review and Recommendations

There is clear evidence to suggest that diabetes does not affect all populations equally. Among adults living with diabetes, those from ethnoracial minority communities—foreign-born, immigrant, refugee, and culturally marginalized—are at increased risk of poor health outcomes. Artificial intelligence (AI) is actively being researched as a means of improving diabetes management and care; however, several factors may predispose AI to ethnoracial bias. To better understand whether diabetes AI interventions are being designed in an ethnoracially equitable manner, we conducted a secondary analysis of 141 articles included in a 2018 review by Contreras and Vehi entitled “Artificial Intelligence for Diabetes Management and Decision Support: Literature Review.” Two members of our research team independently reviewed each article and selected those reporting ethnoracial data for further analysis. Only 10 articles (7.1%) were ultimately selected for secondary analysis in our case study. Of the 131 excluded articles, 118 (90.1%) failed to mention participants’ ethnic or racial backgrounds. The included articles reported ethnoracial data under various categories, including race (n=6), ethnicity (n=2), race/ethnicity (n=3), and percentage of Caucasian participants (n=1). Among articles specifically reporting race, the average distribution was 69.5% White, 17.1% Black, and 3.7% Asian. Only 2 articles reported inclusion of Native American participants. Given the clear ethnic and racial differences in diabetes biomarkers, prevalence, and outcomes, the inclusion of ethnoracial training data is likely to improve the accuracy of predictive models. Such considerations are imperative in AI-based tools, which are predisposed to negative biases due to their black-box nature and proneness to distributional shift. Based on our findings, we propose a short questionnaire to assess ethnoracial equity in research describing AI-based diabetes interventions. At this unprecedented time in history, AI can either mitigate or exacerbate disparities in health care. Future accounts of the infancy of diabetes AI must reflect our early and decisive action to confront ethnoracial inequities before they are coded into our systems and perpetuate the very biases we aim to eliminate. If we take deliberate and meaningful steps now toward training our algorithms to be ethnoracially inclusive, we can architect innovations in diabetes care that are bound by the diverse fabric of our society.

Ability of Current Machine Learning Algorithms to Predict and Detect Hypoglycemia in Patients With Diabetes Mellitus: Meta-analysis

Background

Machine learning (ML) algorithms have been widely introduced to diabetes research including those for the identification of hypoglycemia.

Objective

The objective of this meta-analysis is to assess the current ability of ML algorithms to detect hypoglycemia (ie, alert to hypoglycemia coinciding with its symptoms) or predict hypoglycemia (ie, alert to hypoglycemia before its symptoms have occurred).

Methods

Electronic literature searches (from January 1, 1950, to September 14, 2020) were conducted using the Dialog platform that covers 96 databases of peer-reviewed literature. Included studies had to train the ML algorithm in order to build a model to detect or predict hypoglycemia and test its performance. The set of 2 × 2 data (ie, number of true positives, false positives, true negatives, and false negatives) was pooled with a hierarchical summary receiver operating characteristic model.

Results

A total of 33 studies (14 studies for detecting hypoglycemia and 19 studies for predicting hypoglycemia) were eligible. For detection of hypoglycemia, pooled estimates (95% CI) of sensitivity, specificity, positive likelihood ratio (PLR), and negative likelihood ratio (NLR) were 0.79 (0.75-0.83), 0.80 (0.64-0.91), 8.05 (4.79-13.51), and 0.18 (0.12-0.27), respectively. For prediction of hypoglycemia, pooled estimates (95% CI) were 0.80 (0.72-0.86) for sensitivity, 0.92 (0.87-0.96) for specificity, 10.42 (5.82-18.65) for PLR, and 0.22 (0.15-0.31) for NLR.

Conclusions

Current ML algorithms have insufficient ability to detect ongoing hypoglycemia and considerate ability to predict impeding hypoglycemia in patients with diabetes mellitus using hypoglycemic drugs with regard to diagnostic tests in accordance with the Users’ Guide to Medical Literature (PLR should be ≥5 and NLR should be ≤0.2 for moderate reliability). However, it should be emphasized that the clinical applicability of these ML algorithms should be evaluated according to patients’ risk profiles such as for hypoglycemia and its associated complications (eg, arrhythmia, neuroglycopenia) as well as the average ability of the ML algorithms. Continued research is required to develop more accurate ML algorithms than those that currently exist and to enhance the feasibility of applying ML in clinical settings.

Trial Registration

PROSPERO International Prospective Register of Systematic Reviews CRD42020163682; http://www.crd.york.ac.uk/PROSPERO/display_record.php?ID=CRD42020163682

Artificial Pancreas Control Strategies Used for Type 1 Diabetes Control and Treatment: A Comprehensive Analysis

This paper presents a comprehensive survey about the fundamental components of the artificial pancreas (AP) system including insulin administration and delivery, glucose measurement (GM), and control strategies/algorithms used for type 1 diabetes mellitus (T1DM) treatment and control. Our main focus is on the T1DM that emerges due to pancreas’s failure to produce sufficient insulin due to the loss of beta cells (β-cells). We discuss various insulin administration and delivery methods including physiological methods, open-loop, and closed-loop schemes. Furthermore, we report several factors such as hyperglycemia, hypoglycemia, and many other physical factors that need to be considered while infusing insulin in human body via AP systems. We discuss three prominent control algorithms including proportional-integral- derivative (PID), fuzzy logic, and model predictive, which have been clinically evaluated and have all shown promising results. In addition, linear and non-linear insulin infusion control schemes have been formally discussed. To the best of our knowledge, this is the first work which systematically covers recent developments in the AP components with a solid foundation for future studies in the T1DM field.