Application of Artificial Intelligence in Diabetes Education and Management: Present Status and Promising Prospect

Application of the integrated data platform combined with dietary management for adults with diabetes: A prospective randomized controlled trial

AIMS

To investigate the efficacy of the integrated data platform of cloud hospital combined with dietary management for adults with type 2 diabetes.

MATERIALS AND METHODS

We conducted a randomized controlled clinical trial. One hundred eighty patients with type 2 diabetes were randomly allocated into a control group (Group A) and an experimental group (Group B). Routine standard diabetes care was applied to the patients in Group A. The integrated data platform with dietary management was applied to Group B. Individualized diabetes education videos were sent to the patients through the platform. The primary endpoint was the change in HbA1c and change in body weight from baseline to Week 12 during the follow-up.

RESULTS

At Week 12, HbA1c was 7.4 ± 0.7%, 6.9 ± 0.9% in Groups A and B, P < 0.01. The rate of fasting blood glucose <7 mmol/L, and glycosylated hemoglobin <7% was higher in Group B than in Group A. At Week 12, there was a significant weight loss and body mass index decrease in the overweight or obese patients of the experimental group. Those overweight or obese patients in the experimental group utilizing the appetite suppressant semaglutide achieved the most significant weight loss, with a 13.4% reduction after 12 weeks.

CONCLUSIONS

The integrated data platform combined with personalized diabetes education video delivery was verified to be a more effective management mode for diabetes. For overweight or obese adults with diabetes, the use of semaglutide in conjunction with dietary management and the integrated data platform led to greater weight loss.

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.

Machine Learning and Deep Learning Techniques Applied to Diabetes Research: A Bibliometric Analysis

BACKGROUND

The use of machine learning and deep learning techniques in the research on diabetes has garnered attention in recent times. Nonetheless, few studies offer a thorough picture of the knowledge generation landscape in this field. To address this, a bibliometric analysis of scientific articles published from 2000 to 2022 was conducted to discover global research trends and networks and to emphasize the most prominent countries, institutions, journals, articles, and key topics in this domain.

METHODS

The Scopus database was used to identify and retrieve high-quality scientific documents. The results were classified into categories of detection (covering diagnosis, screening, identification, segmentation, among others), prediction (prognosis, forecasting, estimation), and management (treatment, control, monitoring, education, telemedicine integration). Biblioshiny and RStudio were used to analyze the data.

RESULTS

A total of 1773 articles were collected and analyzed. The number of publications and citations increased substantially since 2012, with a notable increase in the last 3 years. Of the 3 categories considered, detection was the most dominant, followed by prediction and management. Around 53.2% of the total journals started disseminating articles on this subject in 2020. China, India, and the United States were the most productive countries. Although no evidence of outstanding leadership by specific authors was found, the University of California emerged as the most influential institution for the development of scientific production.

CONCLUSION

This is an evolving field that has experienced a rapid increase in productivity, especially over the last years with exponential growth. This trend is expected to continue in the coming years.

Diabetes and artificial intelligence beyond the closed loop: a review of the landscape, promise and challenges

The discourse amongst diabetes specialists and academics regarding technology and artificial intelligence (AI) typically centres around the 10% of people with diabetes who have type 1 diabetes, focusing on glucose sensors, insulin pumps and, increasingly, closed-loop systems. This focus is reflected in conference topics, strategy documents, technology appraisals and funding streams. What is often overlooked is the wider application of data and AI, as demonstrated through published literature and emerging marketplace products, that offers promising avenues for enhanced clinical care, health-service efficiency and cost-effectiveness. This review provides an overview of AI techniques and explores the use and potential of AI and data-driven systems in a broad context, covering all diabetes types, encompassing: (1) patient education and self-management; (2) clinical decision support systems and predictive analytics, including diagnostic support, treatment and screening advice, complications prediction; and (3) the use of multimodal data, such as imaging or genetic data. The review provides a perspective on how data- and AI-driven systems could transform diabetes care in the coming years and how they could be integrated into daily clinical practice. We discuss evidence for benefits and potential harms, and consider existing barriers to scalable adoption, including challenges related to data availability and exchange, health inequality, clinician hesitancy and regulation. Stakeholders, including clinicians, academics, commissioners, policymakers and those with lived experience, must proactively collaborate to realise the potential benefits that AI-supported diabetes care could bring, whilst mitigating risk and navigating the challenges along the way.

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.

Telehealth Technologies in Diabetes Self-management and Education

BACKGROUND

The coronavirus pandemic of 2019 (COVID-19) forced worldwide recognition and implementation of telehealth as a means of providing continuity of care by varied health care institutions. Diabetes is a global health threat with rates that continue to accelerate, thereby causing an increased need for clinicians to provide diabetes care and education to keep up with demand. Utilizing technology to provide education via phone/smartphone, video/audio, web, text message, mobile apps, or a combination of these methods can help improve patient access and clinical outcomes, cut costs, and close gaps in care.

METHODS

While numerous publications have summarized the various tools and technologies available for capturing remote clinical data and their relevance to diabetes care and self-management, this review focuses on self-educational telehealth tools available for diabetes self-management, their advantages and disadvantages, and factors that need to be considered prior to implementation. Recent relevant studies indexed by PubMed were included.

RESULTS

The widespread use and popularity of phones/smartphones, tablets, computers, and the Internet by patients of all age groups, cultures, socioeconomic and geographic areas allow for increased outreach, flexibility, and engagement with diabetes education, either in combination or as an adjunct to traditional in-person visits. Demonstrated benefits of using health technologies for diabetes self-management education include improved lifestyle habits, reduced hemoglobin A1C levels, decreased health care costs, and better medication adherence. Potential drawbacks include lack of regulation, need for staff training on methodologies used, the requirement for patients to be tech savvy, privacy concerns, lag time with technology updates/glitches, and the need for more long-term research data on efficacy.

CONCLUSIONS

Telehealth technologies for diabetes self-education improve overall clinical outcomes and have come a long way. With increasing numbers of patients with diabetes, it is expected that more optimal and user-friendly methodologies will be developed to fully engage and help patients communicate with their physicians.

Exploring the progress of artificial intelligence in managing type 2 diabetes mellitus: a comprehensive review of present innovations and anticipated challenges ahead

A significant worldwide health issue, Type 2 Diabetes Mellitus (T2DM) calls for creative solutions. This in-depth review examines the growing severity of T2DM and the requirement for individualized management approaches. It explores the use of artificial intelligence (AI) in the treatment of diabetes, highlighting its potential for diagnosis, customized treatment plans, and patient self-management. The paper highlights the roles played by AI applications such as expert systems, machine learning algorithms, and deep learning approaches in the identification of retinopathy, the interpretation of clinical guidelines, and prediction models. Examined are difficulties with individualized diabetes treatment, including complex technological issues and patient involvement. The review highlights the revolutionary potential of AI in the management of diabetes and calls for a balanced strategy in which AI supports clinical knowledge. It is crucial to pay attention to ethical issues, data privacy, and joint research initiatives.

Road map for personalized exercise medicine in T2DM

The number of patients with type 2 diabetes mellitus (T2DM) is rising at an alarming rate. Regular physical activity and exercise are cornerstones in the therapy of T2DM. While a one-size-fits-all approach fails to account for many between-subject differences, the use of personalized exercise medicine has the potential of optimizing health outcomes. Here, a road map for personalized exercise therapy targeted at patients with T2DM is presented. It considers secondary complications, glucose management, response heterogeneity, and other relevant factors that might influence the effectiveness of exercise as medicine, taking exercise-medication-diet interactions, as well as feasibility and acceptance into account. Furthermore, the potential of artificial intelligence and machine learning-based applications in assisting sports therapists to find appropriate exercise programs is outlined.

The effect of personalized intelligent digital systems for self-care training on type II diabetes: a systematic review and meta-analysis of clinical trials

AIMS

Type 2 diabetes (T2D) is rising worldwide. Self-care prevents diabetic complications. Lack of knowledge is one reason patients fail at self-care. Intelligent digital health (IDH) solutions have a promising role in training self-care behaviors based on patients' needs. This study reviews the effects of RCTs offering individualized self-care training systems for T2D patients.

METHODS

PubMed, Web of Science, Scopus, Cochrane Library, and Science Direct databases were searched. The included RCTs provided data-driven, individualized self-care training advice for T2D patients. Due to the repeated studies measurements, an all-time-points meta-analysis was conducted to analyze the trends over time. The revised Cochrane risk-of-bias tool (RoB 2.0) was used for quality assessment.

RESULTS

In total, 22 trials met the inclusion criteria, and 19 studies with 3071 participants were included in the meta-analysis. IDH interventions led to a significant reduction of HbA1c level in the intervention group at short-term (in the third month: SMD = - 0.224 with 95% CI - 0.319 to - 0.129, p value < 0.0; in the sixth month: SMD = - 0.548 with 95% CI - 0.860 to - 0.237, p value < 0.05). The difference in HbA1c reduction between groups varied based on patients' age and technological forms of IDH services delivery. The descriptive results confirmed the impact of M-Health technologies in improving HbA1c levels.

CONCLUSIONS

IDH systems had significant and small effects on HbA1c reduction in T2D patients. IDH interventions' impact needs long-term RCTs. This review will help diabetic clinicians, self-care training system developers, and researchers interested in using IDH solutions to empower T2D patients.

The efficacy of theory driven treatment adherence promotion program among type 2 diabetic patients: application of intervention mapping and mHealth

Background

Mobile health interventions (mHealth) may improve health-related lifestyle behaviors and disease management. Successful management of diabetes is patient-centered responsibility. The aim of this research was to determine the efficacy of the theory driven program of promoting treatment adherence in type 2 diabetes (T2DM) patients based on mHealth.

Methods

This quasi-experimental research was conducted on 70 T2DM patients in Tehran, Iran. Participants were randomly divided into intervention (n = 35) or control (n = 35) groups. The data collection tool was a questionnaire based on some of constructs Social Cognitive Theory (SCT) which elicit from formative evaluation. The SCT theory-based intervention program was developed, implemented, and evaluated based on Intervention Mapping (IM) as a framework in 8 sessions using online WhatsApp application. The data was collected through by online interviews before and one month after the implementation of the program and analyzed in SPSS version 16.

Results

After the implementation of the program, a significant increase in self-efficacy (P = 0.009), outcome expectations (P < 0.001), and also diabetes treatment adherence behaviors (P = 0.024) were indicated in the intervention group. The estimated effect sizes for self-efficacy, outcome expectations, social support, and diabetes treatment adherence behaviors were 0.78, 0.06, 0.07, and 0.62, respectively.

Conclusion

Estimated effect size of the implemented intervention was evaluated as "large" effect for diabetes treatment adherence behaviors. Findings indicated the usefulness and efficacy of the mHealth educational program based on SCT constructs and the IM approach in treatment adherence behaviors promotion among T2DM patients in Iran.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40200-023-01291-5.

The Impact of Digital Technology on Self-Management in Cancer: Systematic Review

BACKGROUND

Self-management (SM) plays an important role in supporting patients' adaptation to and management of the symptoms of chronic diseases. Cancer is a chronic disease that requires patients to have responsibility in management. Digital technology has the potential to enhance SM support, but there is little data on what SM skills are most commonly supported by digital technology.

OBJECTIVE

This review aimed to examine the SM core skills that were enabled and supported by digital interventions in people with cancer and identify any predictors of the effect of digital health intervention on SM core skills.

METHODS

Three electronic databases (MEDLINE, Scopus, and CINAHL) were searched for papers, published from January 2010 to February 2022, that reported randomized controlled trials (RCTs) involving patients with cancer or survivors of cancer where a digital technology intervention was evaluated and change in 1 or more SM core skills was a measured outcome.

RESULTS

This systematic review resulted in 12 studies that were eligible to identify which SM core skills were enabled and supported by digital intervention. The total number of participants in the 12 studies was 2627. The most common SM core skills targeted by interventions were decision-making, goal setting, and partnering with health professionals. A total of 8 (67%) out of 12 RCTs demonstrated statistically significant improvement in outcomes including self-efficacy, survivorship care knowledge and attitude, quality of life, increased knowledge of treatment, and emotional and social functioning. A total of 5 (62%) out of 8 positive RCTs used theoretical considerations in their study design; whereas in 1 (25%) out of 4 negative RCTs, theoretical considerations were used. In 3 studies, some factors were identified that were associated with the development of SM core skills, which included younger age (regression coefficient [RC]=-0.06, 95% CI -0.10 to -0.02; P=.002), computer literacy (RC=-0.20, 95% CI -0.37 to -0.03; P=.02), completing cancer treatment (Cohen d=0.31), male sex (SD 0.34 in social functioning; P=.009), higher education (SD 0.19 in social functioning; P=.04), and being a recipient of chemotherapy (SD 0.36 in depression; P=.008). In all 3 studies, there were no shared identical factors that supported the development of SM core skills, whereby each study had a unique set of factors that supported the development of SM core skills.

CONCLUSIONS

Digital technology for patients with cancer appears to improve SM core skills including decision-making, goal setting, and partnering with health care partners. This effect is greater in people who are younger, male, educated, highly computer literate, completing cancer treatment, or a recipient of chemotherapy. Future research should focus on targeting multiple SM core skills and identifying predictors of the effect of digital technology intervention.

TRIAL REGISTRATION

PROSPERO CRD42021221922; https://tinyurl.com/mrx3pfax.

Quantifying postprandial glucose responses using a hybrid modeling approach: Combining mechanistic and data-driven models in The Maastricht Study

Computational models of human glucose homeostasis can provide insight into the physiological processes underlying the observed inter-individual variability in glucose regulation. Modelling approaches ranging from "bottom-up" mechanistic models to "top-down" data-driven techniques have been applied to untangle the complex interactions underlying progressive disturbances in glucose homeostasis. While both approaches offer distinct benefits, a combined approach taking the best of both worlds has yet to be explored. Here, we propose a sequential combination of a mechanistic and a data-driven modeling approach to quantify individuals' glucose and insulin responses to an oral glucose tolerance test, using cross sectional data from 2968 individuals from a large observational prospective population-based cohort, the Maastricht Study. The best predictive performance, measured by R2 and mean squared error of prediction, was achieved with personalized mechanistic models alone. The addition of a data-driven model did not improve predictive performance. The personalized mechanistic models consistently outperformed the data-driven and the combined model approaches, demonstrating the strength and suitability of bottom-up mechanistic models in describing the dynamic glucose and insulin response to oral glucose tolerance tests.

Insulin: evolution of insulin formulations and their application in clinical practice over 100 years

The first preparation of insulin extracted from a pancreas and made suitable for use in humans after purification was achieved 100 years ago in Toronto, an epoch-making achievement, which has ultimately provided a life-giving treatment for millions of people worldwide. The earliest animal-derived formulations were short-acting and contained many impurities that caused adverse reactions, thereby limiting their therapeutic potential. However, since then, insulin production and purification improved with enhanced technologies, along with a full understanding of the insulin molecule structure. The availability of radio-immunoassays contributed to the unravelling of the physiology of glucose homeostasis, ultimately leading to the adoption of rational models of insulin replacement. The introduction of recombinant DNA technologies has since resulted in the era of both rapid- and long-acting human insulin analogues administered via the subcutaneous route which better mimic the physiology of insulin secretion, leading to the modern basal-bolus regimen. These advances, in combination with improved education and technologies for glucose monitoring, enable people with diabetes to better meet individual glycaemic goals with a lower risk of hypoglycaemia. While the prevalence of diabetes continues to rise globally, it is important to recognise the scientific endeavour that has led to insulin remaining the cornerstone of diabetes management, on the centenary of its first successful use in humans.

Artificial intelligence perspective in the future of endocrine diseases

In recent years, artificial intelligence (AI) shows promising results in the diagnosis, prediction, and management of diseases. The move from handwritten medical notes to electronic health records and a huge number of digital data commenced in the era of big data in medicine. AI can improve physician performance and help better clinical decision making which is called augmented intelligence. The methods applied in the research of AI and endocrinology include machine learning, artificial neural networks, and natural language processing. Current research in AI technology is making major efforts to improve decision support systems for patient use. One of the best-known applications of AI in endocrinology was seen in diabetes management, which includes prediction, diagnosis of diabetes complications (measuring microalbuminuria, retinopathy), and glycemic control. AI-related technologies are being found to assist in the diagnosis of other endocrine diseases such as thyroid cancer and osteoporosis. This review attempts to provide insight for the development of prospective for AI with a focus on endocrinology.

Efficacy of Digitally Supported and Real-Time Self-Monitoring of Blood Glucose-Driven Counseling in Patients with Type 2 Diabetes Mellitus: A Real-World, Retrospective Study in North India

PURPOSE

Poor glycemic control is prevalent in patients with type 2 diabetes mellitus (T2DM) in India. This study aims to understand the effectiveness of a smartphone-connected glucometer, real-time feedback, and contextualized counseling on glycemic control and hypoglycemic episodes in T2DM patients.

METHODS

This retrospective, multicenter study reviewed the medical records of T2DM patients belonging to several cities of north India, who were digitally engaged with a smartphone-connected glucometer and who had received at least one counseling session between September 2019 and July 2020. Intervention included self-monitoring of blood glucose (SMBG) using a smartphone-connected glucometer enabled with real-time transmission of information to certified diabetes educators (CDE) and their corresponding counseling based on SMBG findings.

RESULTS

Of 7111 adult T2DM patients included in this study, majority (75%) of the patients received a single session of counseling, and the remaining patients received 2 (16.7%), 3 (5%), 4 (2%), or ≥5 (1.3%) sessions. The mean age of the patients was 51.6 years, and the majority (77.9%) were males. Digital monitoring of BG and counseling with CDE significantly reduced the mean fasting (by 9.6%), pre-prandial (by 9.9%), and post-prandial (by 9.2%) BG values in 53%, 52%, and 54% of patients, respectively. The majority (81.4%) of patients showed no hypoglycemic episode (≤70 mg/dL) post-counseling. The hypoglycemia episodes observed with FBG, pre-prandial, and post-prandial BG values were reduced significantly by 58.5%, 48.1%, and 61.8%, respectively, post-counseling.

CONCLUSION

Digitally supported and real-time SMBG-driven counselling was effective in glycemic control and reduction of hypoglycemic episodes in T2DM patients in India. Moreover, reduction in hypoglycemia may be due to back end real-time support of CDE intervention.

Precision Medicine in Patients with Differential Diabetic Phenotypes: Novel Opportunities from Network Medicine

INTRODUCTION

Diabetes mellitus (DM) comprises differential clinical phenotypes ranging from rare monogenic to common polygenic forms, such as type 1 (T1DM), type 2 (T2DM), and gestational diabetes, which are associated with cardiovascular complications. Also, the high- -risk prediabetic state is rising worldwide, suggesting the urgent need for early personalized strategies to prevent and treat a hyperglycemic state.

OBJECTIVE

We aim to discuss the advantages and challenges of Network Medicine approaches in clarifying disease-specific molecular pathways, which may open novel ways for repurposing approved drugs to reach diabetes precision medicine and personalized therapy.

CONCLUSION

The interactome or protein-protein interactions (PPIs) is a useful tool to identify subtle molecular differences between precise diabetic phenotypes and predict putative novel drugs. Despite being previously unappreciated as T2DM determinants, the growth factor receptor-bound protein 14 (GRB14), calmodulin 2 (CALM2), and protein kinase C-alpha (PRKCA) might have a relevant role in disease pathogenesis. Besides, in silico platforms have suggested that diflunisal, nabumetone, niflumic acid, and valdecoxib may be suitable for the treatment of T1DM; phenoxybenzamine and idazoxan for the treatment of T2DM by improving insulin secretion; and hydroxychloroquine reduce the risk of coronary heart disease (CHD) by counteracting inflammation. Network medicine has the potential to improve precision medicine in diabetes care and enhance personalized therapy. However, only randomized clinical trials will confirm the clinical utility of network- oriented biomarkers and drugs in the management of DM.

Applications of laboratory findings in the prevention, diagnosis, treatment, and monitoring of COVID-19

The worldwide pandemic of coronavirus disease 2019 (COVID-19) presents us with a serious public health crisis. To combat the virus and slow its spread, wider testing is essential. There is a need for more sensitive, specific, and convenient detection methods of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Advanced detection can greatly improve the ability and accuracy of the clinical diagnosis of COVID-19, which is conducive to the early suitable treatment and supports precise prophylaxis. In this article, we combine and present the latest laboratory diagnostic technologies and methods for SARS-CoV-2 to identify the technical characteristics, considerations, biosafety requirements, common problems with testing and interpretation of results, and coping strategies of commonly used testing methods. We highlight the gaps in current diagnostic capacity and propose potential solutions to provide cutting-edge technical support to achieve a more precise diagnosis, treatment, and prevention of COVID-19 and to overcome the difficulties with the normalization of epidemic prevention and control.

Application Prospect of Artificial Intelligence in Rehabilitation and Management of Myasthenia Gravis

Myasthenia gravis (MG) is a chronic autoimmune disease of the nervous system, which is still incurable. In recent years, with the progress of immunosuppressive and supportive treatment, the therapeutic effect of MG in the acute stage is satisfactory, and the mortality rate has been greatly reduced. However, there is still no consensus on how to conduct long-term management of stable MG, such as guiding patients to identify relapses, practice exercise, return to work and school, etc. In the international consensus guidance for management of myasthenia gravis published by the Myasthenia Gravis Foundation of America (MGFA) in 2020, for the first time, "the role of physical training/exercise in MG" was identified as the topic of discussion. Finally, due to a lack of high-quality evidence on physical training/exercise in patients with MG, the topic was excluded after the literature review. Therefore, this paper reviewed the current status of MG rehabilitation research and the difficulties faced by stable MG patients in self-management. It is suggested that we should take advantage of artificial intelligence (AI) and leverage it to develop the data-driven decision support platforms for MG management which can be used for adverse event monitoring, disease education, chronic management, and a wide variety of data collection and analysis.