Search Results (670)

Background: Artificial Intelligence (AI) is increasingly applied in endurance sports to optimize performance, enhance recovery, and personalize nutrition and supplementation. This review synthesizes current knowledge on AI applications in endurance sports, emphasizing implications for metabolic health, nutritional strategies, and recovery optimization, while also addressing ethical considerations and future directions. Methods: A narrative review was conducted using targeted searches of PubMed, Scopus, and Web of Science with cross-referencing. Extracted items included sport/context, data sources, AI methods including machine learning (ML), validation type (internal vs. external/field), performance metrics, comparators, and key limitations to support a structured synthesis; no formal risk-of-bias assessment or meta-analysis was undertaken due to heterogeneity. Results: AI systems effectively integrate multimodal physiological, environmental, and behavioral data to enhance metabolic health monitoring, predict recovery states, and personalize nutrition. Continuous glucose monitoring combined with AI algorithms allows precise carbohydrate management during prolonged events, improving performance outcomes. AI-driven supplementation strategies, informed by genetic polymorphisms and individual metabolic responses, have demonstrated enhanced ergogenic effectiveness. However, significant challenges persist, including measurement validity and reliability of sensor-derived signals and overall dataset quality (e.g., noise, missingness, labeling error), model performance and generalizability, algorithmic transparency, and equitable access. Furthermore, limited generalizability due to homogenous training datasets restricts widespread applicability across diverse athletic populations. Conclusions: The integration of AI in endurance sports offers substantial promise for improving performance, recovery, and nutritional strategies through personalized approaches. Realizing this potential requires addressing existing limitations in model performance and generalizability, ethical transparency, and equitable accessibility. Future research should prioritize diverse, representative, multi-site data collection across sex/gender, age, and race/ethnicity. Coverage should include performance level (elite to recreational), sport discipline, environmental conditions (e.g., heat, altitude), and device platforms (multi-vendor/multi-sensor). Equally important are rigorous external and field validation, transparent and explainable deployment with appropriate governance, and equitable access to ensure scientifically robust, ethically sound, and practically relevant AI solutions. Full article

Background: Diabetes mellitus is a well-established risk factor for surgical site infections (SSIs), particularly periprosthetic joint infections (PJIs) following joint arthroplasty. Although strict glycemic control in the early postoperative period is critical, few studies have evaluated glycemic dynamics using continuous glucose monitoring (CGM) in this setting. This study aimed to characterize early postoperative glycemic patterns using CGM in patients with type 2 diabetes mellitus undergoing lower extremity arthroplasty and to identify factors associated with postoperative hyperglycemia. Methods: We retrospectively analyzed 41 patients with type 2 diabetes who underwent total hip or knee arthroplasty. CGM was used to monitor glucose levels continuously for 48 h after surgery. All patients received standard glycemic management based on a sliding-scale insulin protocol. Patients were classified into two groups: normoglycemia (glucose consistently < 200 mg/dL) and hyperglycemia (glucose ≥ 200 mg/dL at least once within 48 h). Univariable and multivariable logistic regression analyses were conducted to identify predictors of postoperative hyperglycemia. Results: Hyperglycemia occurred in 65.9% of all patients. Univariable analysis identified fasting plasma glucose (FPG), mean postoperative glucose, number of antidiabetic medications, and glucose variability as significant predictors (p < 0.05). In multivariable analysis adjusted for HbA1c, glycoalbumin, and glucose variability, FPG [odds ratio (OR): 1.07; 95% confidence interval (CI): 1.01–1.14; p = 0.024], mean glucose (OR: 1.12; 95% CI: 1.02–1.23; p = 0.017), and glucose variability (OR: 1.19; 95% CI: 1.05–1.35; p = 0.008) remained independently associated with hyperglycemia. Conclusions: CGM revealed a high incidence of early postoperative hyperglycemia despite conventional sliding-scale insulin therapy. These findings highlight the limitations of current glycemic protocols and underscore the potential of CGM as a diagnostic tool to guide individualized glucose management. Future studies should evaluate whether CGM-guided interventions can improve surgical outcomes, particularly in reducing SSI risk among high-risk diabetic patients. Full article

Background/Objectives: Previous studies show that allocating carbohydrates earlier and vegetables/protein later in late-evening meals improves glycemic control in both healthy individuals and those with type 2 diabetes. However, evidence remains insufficient regarding the effects of distributing carbohydrate intake across the day by dividing three regular meals into five smaller meals. Methods: We conducted a randomized, controlled, crossover trial to compare the effects of two dietary patterns: (1) a conventional three-meal pattern with simultaneous intake of all food components, and (2) a five-meal pattern incorporating divided carbohydrate portions and a fixed food order—vegetables first, followed by protein, and then carbohydrates. Eighteen healthy young women consumed the same test meals under both patterns. Glucose fluctuations were monitored using an intermittently continuous glucose monitoring system. Results: The five-meal pattern with food sequencing significantly improved the mean amplitude of glycemic excursions (MAGE; 2.56 ± 0.13 vs. 3.49 ± 0.32 mmol/L, p < 0.01), glucose peak, and incremental area under the glucose curve for breakfast, lunch, and dinner, and the time above the target glucose range [>7.8 mmol/L; 1.4 ± 0.6 vs. 4.2 ± 1.0%, p < 0.01] compared to the three-meal pattern. Conclusions: These findings suggest that divided carbohydrate intake and food order ameliorates the MAGE in healthy young women. Full article

Continuous glucose monitoring (CGM)-based interventions have been predominantly conducted in people with established diabetes. Recently, there has been an increasing interest in using CGM for clinical and research purposes in people without diabetes. In this review, we describe the current evidence regarding the use of CGM in people at high risk of diabetes. To date, there is no strong evidence to support the global implementation of CGM in individuals who are at risk of developing diabetes. However, there are promising results highlighting the benefits of CGM in specific populations such as people living with obesity, prediabetes, gestational diabetes mellitus, metabolic dysfunction-associated steatotic liver disease, other endocrinopathies, and genetic syndromes. Also, CGM has shown promising potential in people with positive islet autoantibodies and pre-symptomatic type 1 diabetes, those treated with medications that induce hyperglycaemia or diabetes, and individuals receiving solid organ transplantation who are at risk of post-transplant diabetes mellitus. However, larger studies are needed to confirm these preliminary results. CGM-derived data are not currently validated for the diagnosis of diabetes. There is no CGM-derived definition of normoglycaemia in people without diabetes. Looking to the future, CGM metrics, in tandem with physical activity, dietary intake, and clinical parameters, and eventually bioinformatics, may inform personalised risk scores for precision prevention of individuals at risk. We conclude that further research is needed to clarify the indications, drawbacks, and feasibility of CGM use in people at high risk of diabetes to identify those groups who could benefit most from this technology. Full article

Introduction: After total pancreatectomy, patients inevitably develop pancreatogenic diabetes with marked glycemic variability and high risk of malnutrition due to both endocrine and exocrine insufficiency. Weight loss and malnutrition can occur even in those with adequate dietary intake and plausible pancreatic enzyme replacement. We hypothesized that glycemic variability is associated with nutritional decline. Methods: We retrospectively analyzed 14 patients who underwent continuous glucose monitoring (CGM) after total pancreatectomy. Nutritional status was assessed using the Geriatric Nutritional Risk Index (GNRI), and patients were classified into malnutrition-risk progression or nutrition-maintaining groups. Then, we evaluated glycemic indices, dietary intake, anthropometry, and pancreatic enzyme replacement therapy (PERT). Results: Insulin use, PERT dose, and dietary intake were approximately comparable between groups. In contrast, the malnutrition-risk progression group showed significantly higher mean glucose and time above range, and lower time in range (TIR). Importantly, TIR consistently showed an inverse association with malnutrition-risk progression across models adjusted for clinical covariates, including time since pancreatectomy, primary diagnosis, insulin regimen, and pancrelipase dose. These findings indicate that the observed relationship between lower TIR and worsening GNRI was independent of dietary intake and adequacy of enzyme replacement therapy, underscoring TIR as a clinically meaningful indicator of nutritional decline in this population. Conclusions: Hyperglycemia and reduced TIR were significantly associated with worsening GNRI after total pancreatectomy, independent of dietary intake or PERT. CGM-based glycemic metrics may help identify patients at risk of malnutrition and guide postoperative management. Full article

Background/Objectives: Accurate blood glucose forecasting is critical for closed-loop insulin delivery systems to support effective disease management in people with type 1 diabetes (T1D). While long short-term memory (LSTM) neural networks have shown strong performance in glucose prediction tasks, the relative performance of individualized versus aggregated training remains underexplored. Methods: In this study, we compared LSTM models trained on individual-specific data to those trained on aggregated data from 25 T1D subjects using the HUPA UCM dataset. Results: Despite having access to substantially less training data, individualized models achieved comparable prediction accuracy to aggregated models, with mean root mean squared error across 25 subjects of 22.52 ± 6.38 mg/dL for the individualized models, 20.50 ± 5.66 mg/dL for the aggregated models, and Clarke error grid Zone A accuracy of 84.07 ± 6.66% vs. 85.09 ± 5.34%, respectively. Subject-level analyses revealed only modest differences between the two approaches, with some individuals benefiting more from personalized training. Conclusions: These findings suggest that accurate and clinically reliable glucose prediction is achievable using personalized models trained on limited individual data, with important implications for adaptive, on-device training, and privacy-preserving applications. Full article

Background and Clinical Significance: Roux-en-Y gastric bypass (RYGB) significantly alters glucose metabolism, yet managing glucose variability in patients undergoing bariatric surgery remains challenging. Continuous Glucose Monitoring (CGM) offers real-time insights into glucose fluctuations and may support long-term metabolic management in this population. This case highlights the utility of CGM in identifying postprandial glycemic variability and guiding dietary adjustments. Case Presentation: A 45-year-old female, 14 years post-RYGB, presented with symptoms including postprandial fatigue, nocturnal cravings, and unexplained weight gain, despite adherence to nutritional guidelines. Her BMI had decreased from 52 kg/m² (pre-surgery) to 29 kg/m². She was provided with a CGM device (FreeStyle Libre 3) by the clinical team and instructed to monitor glucose without modifying her routine initially. Data revealed significant glycemic variability, with peaks up to 220 mg/dL and hypoglycemic dips to 45 mg/dL. Based on this, she adjusted her diet by reducing non-complex carbohydrates and increasing vegetables, proteins, and complex carbohydrates. Within two weeks, her symptoms improved, including better sleep and energy levels, accompanied by a 3 kg weight loss following dietary adjustments informed by CGM feedback. Conclusions: This case suggests how CGM can empower patients having undergone bariatric surgery to manage glucose fluctuations through informed self-management. The patient’s ability to identify and address glucose variability without formal intervention highlights CGM’s potential as a supportive tool in long-term care. While further studies are needed, this case suggests CGM may benefit similar patients experiencing postprandial symptoms or weight regain after bariatric surgery. Full article

This research introduces a hybrid framework that integrates stochastic modeling and machine learning for predicting postprandial glucose levels in individuals with Type 1 Diabetes (T1D). The primary aim is to enhance the accuracy of glucose predictions by merging a biophysical Glucose–Insulin–Meal (GIM) model with advanced machine learning techniques. This framework is tailored to utilize the Kaggle BRIST1D dataset, which comprises real-world data from continuous glucose monitoring (CGM), insulin administration, and meal intake records. The methodology employs the GIM model as a physiological prior to generate simulated glucose and insulin trajectories, which are then utilized as input features for the machine learning (ML) component. For this component, the study leverages the Light Gradient Boosting Machine (LightGBM) due to its efficiency and strong performance with tabular data, while Long Short-Term Memory (LSTM) networks are applied to capture temporal dependencies. Additionally, Bayesian regression is integrated to assess prediction uncertainty. A key advancement of this research is the transition from a deterministic GIM formulation to a stochastic differential equation (SDE) framework, which allows the model to represent the probabilistic range of physiological responses and improves uncertainty management when working with real-world data. The findings reveal that this hybrid methodology enhances both the precision and applicability of glucose predictions by integrating the physiological insights of Glucose Interaction Models (GIM) with the flexibility of data-driven machine learning techniques to accommodate real-world variability. This innovative framework facilitates the creation of robust, transparent, and personalized decision-support systems aimed at improving diabetes management. Full article

This study aims to investigate the effects of continuous glucose monitoring (CGM) on glycemic control in patients with diabetes mellitus (DM) and to identify the sociodemographic or health behavioral factors that influence the outcomes. The data were collected from 510 diabetic patients prescribed to use CGM for 12 weeks and analyzed using SPSS 27.0. Paired samples t-tests were used to compare the glycemic control (HbA1c and fasting glucose) and metabolic health (body mass index and total cholesterol) measures of subjects before and after the CGM use, and independent t-tests were conducted to examine whether the effectiveness of CGM differs according to subjects’ sociodemographic and health behavioral characteristics. As a result of this study, the use of CGM resulted in a significant reduction in HbA1c from 8.09 to 7.48 percent (p < 0.001) and in fasting glucose from 152.41 to 137.16 mg/dL (p < 0.001). In the subgroup analysis of CGM effectiveness, fasting glucose reduction was greater in females than in males and in patients with type 2 diabetes than in those with type 1 diabetes. In conclusion, it is essential to consider patient characteristics to enhance the effectiveness of CGM and to expand its use to type 2 diabetes to reduce the social burden of the disease. Full article

Objective: To explore the lived experiences of type 2 diabetes mellitus (T2DM) patients undergoing peritoneal dialysis (PD) or hemodialysis (HD) using continuous glucose monitoring (CGM). Research Design and Methods: A qualitative phenomenological study was conducted with 50 adult T2DM patients on PD or HD who used CGM for at least 14 days. Semi-structured interviews were audio-recorded and transcribed verbatim. A thematic analysis framework was applied to identify major themes regarding insulin management, CGM utilization, and emotional and social dimensions. Results: Four main themes emerged, each with multiple subthemes. PD patients emphasized enhanced autonomy and frequent insulin adjustments due to dialysate glucose absorption. Conversely, HD patients reported severe post-dialysis fatigue, emotional distress, and limited social engagement often associated with intra-dialytic hypoglycemia. CGM was valued by 85% of participants for improving metabolic awareness and self-management. However, 15% reported barriers such as device cost and technical difficulties. The insights clearly distinguish the differential impact of dialysis modality on daily glucose control and patient well-being. Conclusions: These findings underscore the critical need for patient-centered care incorporating access to CGM and tailored insulin regimens. Equitable implementation of CGM in dialysis settings could significantly enhance glycemic control, emotional resilience, and overall quality of life. Full article

The reliability of health monitoring technologies has become increasingly critical as Ambient Intelligence (AmI) becomes integrated into healthcare. However, a significant gap remains in systematically evaluating reliability, particularly in resource-constrained environments. This study addresses this gap by introducing a comprehensive framework specifically designed to evaluate the reliability of AmI-based health monitoring systems. The proposed framework combines robust simulation-based techniques, including reliability block diagrams (RBDs) and Monte Carlo Markov Chain (MCMC), to evaluate system robustness, data integrity, and adaptability. Validation was performed using real-world continuous glucose monitoring (CGM) and heart rate monitoring (HRM) systems in elderly care. The results demonstrate that the framework successfully identifies critical vulnerabilities, such as rapid initial system degradation and notable connectivity disruptions, and effectively guides targeted interventions that significantly enhance overall system reliability and user trust. The findings contribute actionable insights for practitioners, developers, and policymakers, laying a robust foundation for further advancements in explainable AI, proactive reliability management, and broader applications of AmI technologies in healthcare. Full article

Pregnancy in women with type 1 diabetes mellitus (T1DM) is associated with a high risk of maternal and perinatal complications, and achieving optimal glycaemic control remains a clinical challenge. This article presents a narrative review of the evidence on advanced hybrid closed loop (AHCL) insulin delivery systems in pregnancy, with a focus on maternal glycaemic outcomes, neonatal outcomes, and psychosocial aspects. The relevant literature was identified through a structured search of PubMed, Scopus, and Web of Science (2010–2025), supplemented by guideline documents and reference screening. Eligible studies included randomised controlled trials, observational studies, and qualitative investigations. Data were synthesised thematically. Findings from key trials, including CONCEPTT, AiDAPT, and CRISTAL, demonstrate that AHCL systems improve time in range, lower mean glucose, and reduce hyperglycaemia without increasing hypoglycaemia. Some evidence also suggests improved neonatal outcomes, though statistical significance varies. Qualitative studies highlight reduced anxiety, improved sleep, and enhanced quality of life for women using AHCL during pregnancy. In conclusion, AHCL systems show strong promise in optimising maternal glycaemic control and potentially improving perinatal outcomes. However, larger, unbiased studies and real-world evaluations are needed to confirm their benefits and support broader clinical implementation. Full article

To evaluate the effect of sodium–glucose cotransporter-2 inhibitors (SGLT2i), glucagon-like peptide-1 receptor agonists (GLP-1RAs), and their combination on atrial fibrillation (AF) recurrence after catheter ablation in patients with type 2 diabetes mellitus (T2DM). In a prospective cohort study, patients with T2DM undergoing AF ablation were stratified into three groups: SGLT2i-users, GLP-1RAs-users, and combined SGLT2i/GLP-1RAs users. Diabetics under SGLT2i/GLP-1RAs therapy had worse glycemic control (HbA1c > 7%). AF recurrence was assessed over 12 months using implantable continuous monitoring (ICM). Secondary outcomes included the inflammatory/oxidative stress markers measured at the 12-month follow-up. At the follow-up end, patients treated with SGLT2i/GLP-1RAs versus monotherapy patients showed significantly lower AF recurrence and serum inflammatory/oxidative stress markers, despite having higher HbA1c levels (p < 0.05). Combined SGLT2i/GLP-1RAs therapy reduced AF recurrence following catheter ablation and inflammatory/oxidative stress in T2DM patients. Full article

Background: Continuous glucose monitoring (CGM) has changed the clinical practice in diabetes management during pregnancy; however, its application during caesarean section remains understudied. This feasibility study evaluates the performance, reliability, and clinical utility of two CGM systems—FreeStyle Libre 2 and Medtronic Guardian Connect—during caesarean delivery and the early postpartum period in a patient with pregestational diabetes mellitus (PGDM). Methods: A prospective, single-patient study was conducted. A 32-year-old woman with type 1 diabetes underwent elective caesarean section at 38 weeks of gestation. Both CGM systems were applied over 18 h prior to surgery and monitored continuously through the intraoperative and five-day postpartum period. Glucose data, device performance, and usability were assessed. Results: Both CGM systems provided uninterrupted, high-quality glucose data throughout the perioperative period, including during spinal anaesthesia, surgical manipulation, and postoperative recovery. No sensor displacement nor signal loss occurred. Glycaemic readings remained within the normoglycaemic range (90–100 mg/dL) during surgery, with mild elevations observed during anaesthesia initiation. Postoperatively, both systems showed comparable glucose trends, with slightly lower readings from FreeStyle Libre 2. Conclusions: CGM is feasible and reliable during caesarean section in PGDM patients. These findings support the integration of CGM into obstetric surgical care and highlight the need for larger studies to validate clinical benefits. Full article

Circulating levels of endothelial progenitor cells (EPCs) involved in endothelial homeostasis are often reduced in people with type 1 diabetes (T1D). The Glycemia Risk Index (GRI) quantifies the quality of glucose control by assessing both hypo- and hyperglycemia risk. We aim to investigate the association between the GRI and circulating EPC levels in people with T1D. This cross-sectional study included 132 adults with T1D, on intensive insulin therapy. We calculated GRI from 14 days continuous glucose monitoring-derived metrics and quantified EPCs count by flow cytometry, stratifying results by GRI zones, ranging from A (lowest risk) to E (highest risk). Higher GRI scores were significantly associated with poorer metabolic parameters. Circulating levels of CD34⁺, CD133⁺, KDR⁺, and CD34⁺KDR⁺ cells were lower in participants with a worse GRI compared to adults with a better GRI. Linear regression analyses showed a negative association between GRI and CD34⁺ (β = −1.079, p = 0.006), CD34⁺CD133⁺ (β = −0.581, p = 0.008), and CD34⁺KDR⁺ (β = −0.147, p = 0.010). No significant association was found between HbA_1c and any EPC phenotype. Adults with T1D and a high GRI level had a lower EPCs count. GRI was significantly associated with certain EPC phenotypes, suggesting its potential role as a biomarker for cardiovascular risk assessment. Full article