AIMS:We aimed to determine whether muscular fitness, focussing on power-related components, is reduced in adults with type 1 diabetes (T1D) compared with the group without T1D. METHODS:Adults with T1D and healthy controls were enroled in the study. Muscular fitness was evaluated using Handgrip strength (HG), knee extensor isometric maximal torque (PeakT), and 10-repetition sit-to-stand (STS). Moreover, the specific power index of the lower limb was obtained from the ratio between STS and skeletal muscle mass. RESULTS:Sixty-seven individuals with T1D (32 females, 35 males; age: 38.9 ± 14.6 years) were matched for age, sex, body mass index and physical activity level with a non-diabetic Control Group (CG, n = 67). The T1D group was significantly slower than the control group (CG) during the STS test (20.2 ± 3.8 vs. 18.6 ± 3.3 s, p < 0.01) and exhibited a lower specific power (5.3 ± 1.2 vs. 5.9 ± 1.3 W/kg, p < 0.01). No significant differences were observed in HG and PeakT (p > 0.05) between the T1D and CG groups. CONCLUSIONS:Adults with T1D show a reduction in dynamic muscle function (STS), whereas isometric function appears to be preserved.

Machine Learning techniques are celebrated for their predictive accuracy, uncovering subtle patterns beyond human perception. Among these, Random Forest is a widely adopted ensemble method, valued for its robust performance and ease of use, especially in scenarios where the cost of errors is significant. However, the inherent opacity of such models raises concerns about their interpretability and trustworthiness. In this study, we apply the Random Forest algorithm to analyze data from the Italian National Institute of Statistics (Istituto Nazionale di Statistica, Istat), specifically the European Health Interview Survey (EHIS), to identify risk factors associated with depression in Italy. To enhance transparency and interpretability, we subsequently employ the Explainable Ensemble Trees methodology to explore and understand the decision-making processes of the Random Forest model. The goal is to demonstrate how E2Tree can provide actionable insights for targeted interventions, supporting public health efforts in addressing mental health challenges.

Today, there is a growing demand for models that not only predict outcomes, but also explain the decision-making processes involved. This is particularly true in domains where decisions have significant consequences, such as in the financial and health care sectors. Whether diagnosing patients, approving loans, or managing public health initiatives, decisions influenced by machine learning methods models can have profound implications. To address this explainability issue, we show the predictive power of ensemble decision trees with the transparent logic of single trees, providing deep understanding of ensemble reasoning.

Background: Longitudinal Displacement (LD) is the relative motion of the intima-media upon adventitia of the arterial wall during the cardiac cycle, probably linked to atherosclerosis. It has a direction, physiologically first backward in its main components with respect to the arterial flow. Here, LD was investigated in various disease and in presence of a unilateral carotid stent. Methods: Carotid acquisitions were performed by ultrasound imaging on both body sides of 75 participants (150 Arteries). LD was measured in its percent quantity and direction. Results: Obesity (p = 0.001) and carotid plaques (p = 0.01) were independently associated to quantity decrease of LD in the whole population. In a subgroup analysis, it was respectively 143% in healthy (n = 48 carotids), 129% (n = 34) in presence of cardiovascular risk factors, 121% (n = 20) in MACE patients, 119% (n = 24) in the carotid contralateral to a stent, 110% (n = 24) in carotids with stents. Regarding the direction of LD, in a subgroup analysis an inverted movement was identified in aged (p = 0.001) and diseased (p = 0.001) participants who also showed less quantity of LD (p = 0.001), but independently with age only (p = 0.002) in the whole population. Conclusions: This observational study suggests that LD within carotid wall layers is lower additively with ageing, cardiovascular risk factors, cardiovascular diseases, and stent. Even if stent is surely beneficial, these data might shed some light on stent restenosis, emphasising the need for interventional studies.

Machine Learning methods have gained significant attention for their ability to deliver high predictive accuracy and reveal complex, non-obvious patterns in data. Among these, Random Forest stands out as a popular ensemble technique, appreciated for its robustness and practical applicability, particularly in contexts where minimizing prediction errors is crucial. Nevertheless, the inherent complexity and lack of transparency in such models often limit their interpretability and can hinder user trust. In this study, we employ the Random Forest algorithm to analyze data from the European Health Interview Survey (EHIS), conducted by the Italian National Institute of Statistics (ISTAT), with the aim of identifying key risk factors associated with depression in Italy. To improve the interpretability of the model's output, we apply the Explainable Ensemble Trees approach, which allows for a deeper understanding of the internal decision-making mechanisms of Random Forest. The aim is to shed light on the determinants of mental health conditions, providing valuable insights to better understand the factors contributing to the risk of depression.

INTRODUCTION:Diabetes is a chronic disease with high prevalence, necessitating advanced technology to achieve glycemic targets and reduce complications. Continuous glucose monitoring (CGM) has become a cornerstone in diabetes management, with the Freestyle Libre (FSL) systems being some of the most widely used devices. AREAS COVERED:This review focuses on FSL systems, each including an all-in-one sensor and transmitter, a handheld reader and a Mobile Medical App (MMA). Glucose data are uploaded to a dedicated cloud-based platform for analysis. Over the years, FSL has evolved with new features offering valuable support for individuals with diabetes, caregivers, and healthcare providers. Clinical trials and real-world studies have demonstrated efficacy and safety across diverse populations, including individuals with type 1 and type 2 diabetes, adolescents, and pregnant women with diabetes. EXPERT OPINION:FSL is a user-friendly system that meets the needs of patients and healthcare providers. The MMA allows a review of glucose metrics and pattern identification, and supports educational strategies and patient-tailored treatment. Future advancements, including ketone monitoring, integration with wearables and other devices, and telemedicine applications, will further optimize diabetes care and prevention.

Plasma Volume Status (PVS), an index that quantifies the deviation of an individual’s plasma volume from the expected volume, plays an important role in cardiovascular homeostasis. Although PVS expansion is well recognized in conditions like heart failure and chronic kidney disease, its relationship with type 2 diabetes (T2D) and antidiabetic treatment remains uncertain. We conducted an observational study comparing PVS in adults with T2D and healthy controls matched for sex, age, and BMI. In a separate analysis of a larger T2D cohort, we investigated the association between PVS and different antidiabetic therapies. PVS was calculated using established formulas incorporating hematocrit, body weight, and sex. PVS was significantly expanded in individuals with T2D compared with healthy controls, with a mean difference of + 3.73

BACKGROUND AND AIMS:Insulin resistance is a growing feature in type 1 diabetes (T1D). It can be quantified by calculating the estimated glucose disposal rate (eGDR) with the Epstein's formula, which includes laboratory-measured glycated hemoglobin (HbA1c). We aimed the current research to assess the agreement between the conventional eGDR formula and an alternative one (eGDR-GMI) incorporating the glucose management indicator (GMI) derived from continuous glucose monitoring (CGM). We also explored the relationship between eGDR-GMI, cardiovascular risk factors, and the prevalence of diabetes-related complications. METHODS AND RESULTS:We designed a cross-sectional study that included adults with T1D. eGDR-GMI and eGDR (mg/kg/min) were calculated using GMI or HbA1c, waist circumference, and hypertensive state. Clinical data were collected from electronic medical records. The analyses encompassed 158 participants with a mean age of 39 ± 13 years. The Bland-Altman analysis showed a good agreement between eGDR-GMI and eGDR. When we divided participants in eGDR-GMI tertiles we found a higher prevalence of diabetes-related complications and a less favorable metabolic profile in the lowest eGDR-GMI tertile. The relative risk of retinopathy, nephropathy, and neuropathy significantly increased by approximately 1 unit with each decrease in eGDR-GMI, regardless of age, sex, disease duration, lipids, and smoking habit. CONCLUSIONS:eGDR-GMI represents a valid and robust alternative to the eGDR to assess insulin resistance in T1D. Low eGDR-GMI is associated with diabetes complications and a less favorable metabolic profile. Incorporating the eGDR-GMI into clinical practice can enhance the characterization of T1D people and allow for a more personalized treatment approach.

Aims: The study aimed to evaluate blood flow (BF) and microvascular function in the forearm of people with type 1 and type 2 diabetes at rest and after ischemia. Microvascular function plays a crucial role in regulating BF in peripheral tissues based on metabolic demand. Methods: People with diabetes and sex-matched healthy controls were recruited. Brachial artery diameter and blood velocity were continuously measured at rest and after ischemia by an automatic tracking system. BF and vascular conductance were then calculated. Results: Forty-nine people with diabetes and 49 controls were enrolled. BF at rest and after ischemia was significantly higher in people with diabetes than controls: Type 1, 243 +/- 116 and 631 +/- 233 ml/min; controls, 180 +/- 106 and 486 +/- 227 ml/min; Type 2, 332 +/- 149 and 875 +/- 293 ml/min; controls 222 +/- 106 and 514 +/- 224 ml/min. Vascular conductance was significantly higher in Type 2 than in controls at rest and after ischemia. Conclusions: People with diabetes exhibited significantly increased BF, with Type 2 also showing heightened vascular conductance. Activating metabolic pathways triggered by hyperglycemia may lead to distinct vascular redistribution, potentially impairing blood flow over time. These findings of the study underscore the importance of understanding overall vascular dynamics in diabetes and its implications for vascular health.

AIMS:Individuals with type 1 diabetes (T1D) do not appear to have an elevated risk of severe Coronavirus Disease 19 (COVID-19). Pre-existing immune reactivity to Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) in unexposed individuals may serve as a protective factor. Hence, our study was designed to evaluate the existence of T cells with reactivity against SARS-CoV-2 antigens in unexposed patients with T1D. MATERIALS AND METHODS:Peripheral blood mononuclear cells (PBMCs) were collected from SARS-CoV-2 unexposed patients with T1D and healthy control subjects. SARS-CoV-2 specific T cells were identified in PBMCs by ex-vivo interferon (IFN)γ-ELISpot and flow cytometric assays. The epitope specificity of T cells in T1D was inferred through T Cell Receptor sequencing and GLIPH2 clustering analysis. RESULTS:T1D patients unexposed to SARS-CoV-2 exhibited higher rates of virus-specific T cells than controls. The T cells primarily responded to peptides from the ORF7/8, ORF3a, and nucleocapsid proteins. Nucleocapsid peptides predominantly indicated a CD4+ response, whereas ORF3a and ORF7/8 peptides elicited both CD4+ and CD8+ responses. The GLIPH2 clustering analysis of TCRβ sequences suggested that TCRβ clusters, associated with the autoantigens proinsulin and Zinc transporter 8 (ZnT-8), might share specificity towards ORF7b and ORF3a viral epitopes. Notably, PBMCs from three T1D patients exhibited T cell reactivity against both ORF7b/ORF3a viral epitopes and proinsulin/ZnT-8 autoantigens. CONCLUSIONS:The increased frequency of SAR-CoV-2- reactive T cells in T1D patients might protect against severe COVID-19 and overt infections. These results emphasise the long-standing association between viral infections and T1D.

Diastolic dysfunction represents the earliest and most common manifestation of diabetic cardiomyopathy. Nitric oxide (NO), a potent vasodilator and anti-inflammatory mediator released from the subendocardial and coronary endothelium, favors left ventricular distensibility and relaxation. In type 2 diabetes (T2D), the NO bioavailability is reduced due to the oxidative stress and inflammatory state of the endothelium, because of chronic hyperglycemia. The aim of the present research is to evaluate the relationship between endothelial function and diastolic function in subjects with T2D. Subjects with T2D and age and sex-matched healthy controls were consecutively recruited. All participants underwent flow-mediated dilation (FMD) to assess endothelial function, and echocardiography to evaluate diastolic function. Thirty-five patients (6 women, 29 men) and 35 healthy controls were included in the final analysis. FMD was significantly lower in T2D than controls (4.4 ± 3.4 vs. 8.5 ± 4.3

Background: Type 1 Diabetes Mellitus (T1DM) is a chronic metabolic disease affecting millions of people worldwide. T1DM requires patients to continuously monitor their blood glucose levels. Due to pancreatic dysfunctions, patients use insulin injections to correct glucose values by synthetic insulin. Continuous Glucose Monitoring (CGM) is a system which includes an algorithm allowing to measure (and in some cases to predict) glucose levels at a frequent sampling time. This enable implementing advanced devices, including automated insulin pump delivery. Nevertheless, CGM still presents some limitations, including (i) the delay (time lag) in detecting change in glucose levels compared to the traditional blood glucose measurement, and (ii) the lack of a sufficient and acceptable time to accurately predict glucose values. Methods: We propose a framework based on a Gated Recurrent Unit (GRU) model to forecast both short- and long-term glucose values using heart rate (HR) and interstitial glucose (IG) values. The framework acquires HR and IG data and predicts glucose values with higher precision compared to state-of-the-art models. For training and testing the proposed framework, we used the OhioT1DM Dataset, which includes physiological data such as HR and IG values collected over an 8-week observation period. Additionally, we validated our framework using two other glucose datasets to ensure its generalizability across different HR and IG sampling frequencies. The proposed framework can be used to optimize the CGM system by incorporating patient HR measurements, thereby improving the prediction of short- and long-term glucose levels and reducing risks associated with conditions like hypoglycemia. Results: Experimental tests were conducted using HR and IG data from the OhioT1DM Dataset, as well as from two additional T1DM patient datasets. We analyzed 6 patients from Ohio dataset while we validated the algorithm on 23 patients coming from two different university hospitals (6 from the University of Catanzaro medical hospital and 17 gathered from a validated study at IRCCS San Matteo Hospital in Pavia) for a total number of 29 patients. Our framework demonstrates an improvement in forecasting IG values in terms of RMSE and MAE for different choice of prediction horizons (PH). In the case of a PH of 5, 10, 20, 30, and 60 min, we reach an RMSE of 5.0, 9.38, 15.27, 20.48, and 34.16 respectively. The framework is freely available as an open-source, with an example dataset on a GitHub repository (see https://github.com/rafgia/attention_to_glycemia). Conclusion: Our framework offers a promising solution for improving glucose level prediction and management in T1DM patients. By leveraging a GRU model and incorporating HR and IG values, we achieve more precise glucose level forecasting compared to state-of-the-art models. This approach not only enhances the accuracy of glucose predictions but also mitigates the risks associated with hypoglycemia.

Diabetic nephropathy (DN) is a serious complication of type 2 diabetes (T2D), necessitating early risk assessment and monitoring to guide intervention strategies. This study explores an integrated machine learning approach to identify potential risk predictors of this complication. Using a clinical dataset of diabetic patients, we developed a model for predicting the risk of occurrence of diabetic nephropathy in patients with T2D from electronic medical records (EMR). The results show that the support vector machine (SVM) algorithm with Gaussian kernel can predict the risk of developing DN from 3 to 8 years in patients with T2D with an accuracy of about 75% and that a patient with high levels of body mass index (BMI), white blood cells (WBC) count, creatininemia, triglyceride-glucose index (TyG), glycated hemoglobin (HbA1c), potential advanced diabetic condition with comorbidities (such as hypertension and cardiac diseases), low levels of hemoglobin, estimated glomerular filtration rate (eGFR), indirect bilirubin at baseline, with longer duration of diabetes and with past or current smoking habits represent potential risk factors in the onset of this pathology at a distance of 3 to 8 years from the visit. This work provides the basis for studying the dataset under investigation and then enriching it with features produced by diabetes-specific mathematical models (MMs) to obtain a hybrid model (AI/MM) that can improve the predictive model.

We evaluated the acute effects of yoga compared to cycling on glucose change and variability, and the occurrence of hypoglycemia in adults with type 1 diabetes. Fifteen participants performed 50 min of cycling or yoga. Glucose values were collected before and after exercise. Coefficient of variation (CV) and hypoglycemic episodes were evaluated from the start up to 12 h after exercise. Cycling and yoga significantly reduced glucose values during exercise, and CV was lower after yoga. One hypoglycemic episode occurred with yoga and seven with cycling. Yoga is a safe exercise that acutely reduces glucose values, but with lower risk of hypoglycemia compared to cycling. (c) 2024 Sports Medicine Australia. Published by Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

Optical coherence tomography angiography (OCTA) is an innovative and reliable technique detecting the early preclinical retinal vascular change in patients with diabetes. We have designed our study to evaluate whether an independent relationship exists between continuous glucose monitoring (CGM)-derived glucose metrics and OCTA parameters in young adult patients with type 1 diabetes without diabetic retinopathy (DR). Inclusion criteria were age ≥ 18 years, diagnosis of type 1 diabetes from ≥ 1 year, stable insulin treatment in the last three months, use of real-time CGM, and CGM wear time ≥ 70%. Each patient underwent dilated slit lamp fundus biomicroscopy to exclude the presence of DR. A skilled operator performed OCTA scans in the morning to avoid possible diurnal variation. CGM-derived glucose metrics from the last 2 weeks were collected through the dedicated software during OCTA. Forty-nine patients with type 1 diabetes (age 29 [18; 39] years, HbA1c 7.7 ± 1.0%) and 34 control subjects participated in the study. Vessel density (VD) of the whole image and parafoveal retina in the superficial (SCP) and deep capillary plexus (DCP) was significantly lower in patients with type 1 diabetes compared to controls. The coefficient of variation of average daily glucose, evaluated by CGM, significantly correlated with foveal and parafoveal VD in SCP and with foveal VD in DCP. High glucose variability might be responsible for the early increase of VD in these areas. Prospective studies may help understand if this pattern precedes DR. The difference we detected between patients with and without diabetes confirms that OCTA is a reliable tool for detecting early retinal abnormalities.