Machine Learning Models for Prediction of Diabetic Microvascular Complications.

IMPORTANCE AND AIMS: Diabetic microvascular complications significantly impact morbidity and mortality. This review focuses on machine learning/artificial intelligence (ML/AI) in predicting diabetic retinopathy (DR), diabetic kidney disease (DKD), and diabetic neuropathy (DN). METHODS: A comprehensive PubMed search from 1990 to 2023 identified studies on ML/AI models for diabetic microvascular complications. The review analyzed study design, cohorts, predictors, ML techniques, prediction horizon, and performance metrics. RESULTS: Among the 74 identified studies, 256 featured internally validated ML models and 124 had externally validated models, with about half being retrospective. Since 2010, there has been a rise in the use of ML for predicting microvascular complications, mainly driven by DKD research across 27 countries. A more modest increase in ML research on DR and DN was observed, with publications from fewer countries. For all microvascular complications, predictive models achieved a mean (standard deviation) c-statistic of 0.79 (0.09) on internal validation and 0.72 (0.12) on external validation. Diabetic kidney disease models had the highest discrimination, with c-statistics of 0.81 (0.09) on internal validation and 0.74 (0.13) on external validation, respectively. Few studies externally validated prediction of DN. The prediction horizon, outcome definitions, number and type of predictors, and ML technique significantly influenced model performance. CONCLUSIONS AND RELEVANCE: There is growing global interest in using ML for predicting diabetic microvascular complications. Research on DKD is the most advanced in terms of publication volume and overall prediction performance. Both DR and DN require more research. External validation and adherence to recommended guidelines are crucial.

Effectiveness of artificial intelligence vs. human coaching in diabetes prevention: a study protocol for a randomized controlled trial.

BACKGROUND: Prediabetes is a highly prevalent condition that heralds an increased risk of progression to type 2 diabetes, along with associated microvascular and macrovascular complications. The Diabetes Prevention Program (DPP) is an established effective intervention for diabetes prevention. However, participation in this 12-month lifestyle change program has historically been low. Digital DPPs have emerged as a scalable alternative, accessible asynchronously and recognized by the Centers for Disease Control and Prevention (CDC). Yet, most digital programs still incorporate human coaching, potentially limiting scalability. Furthermore, existing effectiveness results of digital DPPs are primarily derived from per protocol, longitudinal non-randomized studies, or comparisons to control groups that do not represent the standard of care DPP. The potential of an AI-powered DPP as an alternative to the DPP is yet to be investigated. We propose a randomized controlled trial (RCT) to directly compare these two approaches. METHODS: This open-label, multicenter, non-inferiority RCT will compare the effectiveness of a fully automated AI-powered digital DPP (ai-DPP) with a standard of care human coach-based DPP (h-DPP). A total of 368 participants with elevated body mass index (BMI) and prediabetes will be randomized equally to the ai-DPP (smartphone app and Bluetooth-enabled body weight scale) or h-DPP (referral to a CDC recognized DPP). The primary endpoint, assessed at 12 months, is the achievement of the CDC's benchmark for type 2 diabetes risk reduction, defined as any of the following: at least 5% weight loss, at least 4% weight loss and at least 150 min per week on average of physical activity, or at least a 0.2-point reduction in hemoglobin A1C. Physical activity will be objectively measured using serial actigraphy at baseline and at 1-month intervals throughout the trial. Secondary endpoints, evaluated at 6 and 12 months, will include changes in A1C, weight, physical activity measures, program engagement, and cost-effectiveness. Participants include adults aged 18-75 years with laboratory confirmed prediabetes, a BMI of ≥ 25 kg/m2 (≥ 23 kg/m2 for Asians), English proficiency, and smartphone users. This U.S. study is conducted at Johns Hopkins Medicine in Baltimore, MD, and Reading Hospital (Tower Health) in Reading, PA. DISCUSSION: Prediabetes is a significant public health issue, necessitating scalable interventions for the millions affected. Our pragmatic clinical trial is unique in directly comparing a fully automated AI-powered approach without direct human coach interaction. If proven effective, it could be a scalable, cost-effective strategy. This trial will offer vital insights into both AI and human coach-based behavioral change strategies in real-world clinical settings. TRIAL REGISTRATION: ClinicalTrials.gov NCT05056376. Registered on September 24, 2021, https://clinicaltrials.gov/study/NCT05056376.

A Retrospective Cohort Analysis of Two Computerized Insulin Infusion Protocols.

OBJECTIVE: The primary objective of this analysis was to compare the safety and efficacy of a novel computerized insulin infusion protocol (CIIP), the Lalani Insulin Infusion Protocol (LIIP), with an established CIIP, Glucommander. METHODS: We conducted a 10-month retrospective analysis of 778 patients in whom LIIP was used (August 18, 2020 to June 25, 2021) at six HonorHealth Hospitals in the Phoenix metropolitan area. These data were compared with Glucommander that was used at those same hospitals from January 1, 2018 to August 17, 2020, n = 4700. Primary end points of the project included average time to euglycemia and average time in hyperglycemia (>180 mg/dL) and hypoglycemia (<70 mg/dL). Additional subgroup analysis was done to evaluate CIIP performance in patients in whom maintenance of euglycemia was more challenging. RESULTS: The LIIP had a faster time to euglycemia (191 vs 222 minutes, P < .001) and similar time in hypoglycemia (2.79 vs 2.76 minutes, P = .50) for all patients, when compared with Glucommander. Similar observations were made for the following subgroups: diabetic ketoacidosis/hyperosmolar hyperglycemic state (DKA/HHS) patients, COVID-19 patients, patients on steroids, patients with ≥60 glomerular filtration rate (GFR), patients with renal insufficiency, and patients with sepsis. CONCLUSIONS: The LIIP is a safe and effective CIIP in managing intravenous insulin infusion rates. Utilization of LIIP resulted in reduced time to euglycemia, P < .001, when compared with Glucommander and did not cause increased hypoglycemia during the project period. Contributing factors to the success of LIIP may include improved clinical workflow, learnability and ease of use, compatibility with the Epic electronic health record (EHR), and its unique, dynamic and adaptive algorithm.

Decreased IGF1R attenuates senescence and improves function in pancreatic ß-cells.

Introduction: The enhanced ß-cell senescence that accompanies insulin resistance and aging contributes to cellular dysfunction and loss of transcriptional identity leading to type 2 diabetes (T2D). While senescence is among the 12 recognized hallmarks of aging, its relation to other hallmarks including altered nutrient sensing (insulin/IGF1 pathway) in ß-cells is not fully understood. We previously reported that an increased expression of IGF1R in mouse and human ß-cells is a marker of older ß-cells; however, its contribution to age-related dysfunction and cellular senescence remains to be determined. Methods: In this study, we explored the direct role of IGF1R in ß-cell function and senescence using two independent mouse models with decreased IGF1/IGF1R signaling: a) Ames Dwarf mice (Dwarf +/+), which lack growth hormone and therefore have reduced circulating levels of IGF1, and b) inducible ß-cell-specific IGF1R knockdown (ßIgf1rKD) mice. Results: Compared to Dwarf+/- mice, Dwarf+/+ mice had lower body and pancreas weight, lower circulating IGF1 and insulin levels, and lower IGF1R and p21Cip1 protein expression in ß-cells, suggesting the suppression of senescence. Adult ßIgf1rKD mice showed improved glucose clearance and glucose-induced insulin secretion, accompanied by decreased p21Cip1 protein expression in ß-cells. RNA-Seq of islets isolated from these ßIgf1rKD mice revealed the restoration of three signaling pathways known to be downregulated by aging: sulfide oxidation, autophagy, and mTOR signaling. Additionally, deletion of IGF1R in mouse ß-cells increased transcription of genes important for maintaining ß-cell identity and function, such as Mafa, Nkx6.1, and Kcnj11, while decreasing senescence-related genes, such as Cdkn2a, Il1b, and Serpine 1. Decreased senescence and improved insulin-secretory function of ß-cells were also evident when the ßIgf1rKD mice were fed a high-fat diet (HFD; 60% kcal from fat, for 5 weeks). Discussion: These results suggest that IGF1R signaling plays a causal role in aging-induced ß-cell dysfunction. Our data also demonstrate a relationship between decreased IGF1R signaling and suppressed cellular senescence in pancreatic ß-cells. Future studies can further our understanding of the interaction between senescence and aging, developing interventions that restore ß-cell function and identity, therefore preventing the progression to T2D.