Search Results (1,010)

Objective: This study aimed to identify plasma exosomal microRNAs (miRNAs) associated with weight loss and type 2 diabetes (T2D) remission following low-calorie diet (LCD) intervention. Methods: A 6-month dietary intervention targeting T2D remission was conducted among individuals with T2D. Participants underwent a 3-month intensive weight loss phase consuming LCD (815–835 kcal/day) and a 3-month weight maintenance phase (N = 32). Sixteen participants were randomly selected for characterization of plasma-derived exosomal miRNA profiles at baseline, 3 months, and 6 months using small RNA sequencing. Linear mixed-effects models were used to identify differentially expressed exosomal miRNAs between responders and non-responders. Pathway enrichment analyses were conducted using target mRNAs of differentially expressed miRNAs. Logistic regression models assessed the predictive value of differentially expressed miRNAs for T2D remission. Results: Among the 16 participants, 6 achieved weight loss ≥10% and 12 achieved T2D remission. Eighteen exosomal miRNAs, including miR-92b-3p, miR-495-3p, and miR-452b-5p, were significantly associated with T2D remission and weight loss. Pathway analyses revealed enrichment in PI3K-Akt pathway, FoxO signaling pathway, and insulin receptor binding. The addition of individual miRNAs including miR-15b-3p, miR-26a-5p, and miR-3913-5p to base model improved the area under the curve values by 0.02–0.08 at 3 months and by 0.02–0.06 at 6 months for T2D remission. Conclusions: This study identified exosomal miRNAs associated with T2D remission and weight loss following LCD intervention. Several exosomal miRNAs might serve as valuable predictors of T2D remission in response to LCD intervention. Full article

Cardiovascular diseases (CVDs) remain the leading cause of death globally, underscoring the urgent need for data-driven early diagnostic tools. This study proposes a multilayer artificial neural network (ANN) model for heart disease prediction, developed using a real-world clinical dataset comprising 13,981 patient records. Implemented on the Orange data mining platform, the ANN was trained using backpropagation and validated through 10-fold cross-validation. Dimensionality reduction via principal component analysis (PCA) enhanced computational efficiency, while Shapley additive explanations (SHAP) were used to interpret model outputs. Despite achieving 83.4% accuracy and high specificity, the model exhibited poor sensitivity to disease cases, identifying only 76 of 2233 positive samples, with a Matthews correlation coefficient (MCC) of 0.058. Comparative benchmarks showed that random forest and support vector machines significantly outperformed the ANN in terms of discrimination (AUC up to 91.6%). SHAP analysis revealed serum creatinine, diabetes, and hemoglobin levels to be the dominant predictors. To address the current study’s limitations, future work will explore LIME, Grad-CAM, and ensemble techniques like XGBoost to improve interpretability and balance. This research emphasizes the importance of explainability, data representativeness, and robust evaluation in the development of clinically reliable AI tools for heart disease detection. Full article

More than 14% of the world’s population suffered from diabetes mellitus in 2022. This metabolic condition is defined by increased blood glucose concentrations. Among the different types of diabetes, type 1 diabetes, caused by a lack of insulin secretion, is particularly challenging to treat. In this regard, automatic glucose level estimation implements Continuous Glucose Monitoring (CGM) devices, showing positive therapeutic outcomes. AI-based glucose prediction has commonly followed a deterministic approach, usually with a lack of interpretability. Therefore, these AI-based methods do not provide enough information in critical decision-making scenarios, like in the medical field. This work intends to provide accurate, interpretable, and personalized glucose prediction using the Temporal Fusion Transformer (TFT), and also includes an uncertainty estimation. The TFT was trained using two databases, an in-house-collected dataset and the OhioT1DM dataset, commonly used for glucose forecasting benchmarking. For both datasets, the set of input features to train the model was varied to assess their impact on model interpretability and prediction performance. Models were evaluated using common prediction metrics, diabetes-specific metrics, uncertainty estimation, and interpretability of the model, including feature importance and attention. The obtained results showed that TFT outperforms existing methods in terms of RMSE by at least 13% for both datasets. Full article

Background: Type 2 diabetes (T2D) is a major global health concern influenced by sociodemographic and lifestyle factors. This study compared T2D risk scores between commerce and industry sectors and assessed the associations of age, sex, education, physical activity, diet, and smoking with elevated risk. Methods: This cross-sectional study included 56,856 men and 12,872 women employed in the commerce (n = 27,448) and industry (n = 42,280) sectors across Spain. Anthropometric, clinical, and biochemical data were collected. Four validated T2D risk scores (QDscore, Finrisk, Canrisk, and TRAQ-D) were calculated. Multinomial logistic regression models estimated adjusted odds ratios (ORs) for high-risk categories by sociodemographic and lifestyle characteristics. Results: Women in the industrial sector had significantly higher age, BMI, waist circumference, and lipid levels than those in commerce; differences among men were less marked. Across all participants, higher T2D risk scores were independently associated with physical inactivity (OR up to 12.49), poor Mediterranean diet adherence (OR up to 6.62), industrial employment (OR up to 1.98), and older age. Male sex was strongly associated with high Canrisk scores (OR = 6.31; 95% CI: 5.12–7.51). Conclusions: Employment in the industrial sector, combined with sedentary behavior and poor dietary habits, is independently associated with higher predicted T2D risk. Workplace prevention strategies should prioritize multicomponent interventions targeting modifiable risk factors, especially in high-risk subgroups such as older, less-educated, and inactive workers. Full article

Metformin is one of the most commonly used medications to treat type 2 diabetes. In addition to lowering blood sugar, it can also promote the regeneration of certain organs or tissues. Planarian Dugesia japonica, with its remarkable regenerative capacity, has become an important model organism for studying pharmacology and regenerative medicine. Planarian eyespot regeneration involves precise tissue regeneration via mechanisms like cell proliferation, differentiation, and gene regulation following body damage. Experiments on planarian eyespot regeneration have confirmed that 1 mM metformin significantly promotes regeneration. Through analysis of the regenerating planarian miRNA database and the metformin-treated transcriptome database, combined with target gene prediction by TargetScan, the DjmiR-27b/DjPax6 axis was finally determined as the research focus. qPCR showed that metformin significantly affects the expression levels of DjmiR-27b and DjPax6. DjPax6 was identified as the target gene of DjmiR-27b through dual luciferase reporter gene analysis. Functional experiments revealed that metformin regulates the expression of DjPax6 via DjmiR-27b, thereby influencing the regeneration of planarian eyespots. In situ hybridization showed that both DjmiR-27b and DjPax6 are expressed throughout the entire body. This study reveals the molecular mechanism of metformin regulating planarian regeneration through miRNA, providing further insights into its role in the field of regeneration. Full article

Abnormalities in plasma lipoproteins observed in patients with diabetes promote atherosclerosis. However, the association between various lipid species and classes and atherosclerosis remains unclear. Here, we aimed to identify the plasma lipid characteristics associated with atherosclerosis progression in patients with diabetes. We performed semi-targeted lipidomic analysis of fasting plasma samples using supercritical fluid chromatography coupled with mass spectrometry in two independent patient groups with type 2 diabetes (n = 223 and 31) and evaluated cross-sectional associations between plasma lipids and carotid intima-media thickness (CIMT). Ten plasma lipid species, including eight diacylglycerols (DGs), and total DG levels were significantly associated with CIMT in both groups. Patients of the former group were partly observed for 5 years, and we investigated associations between DGs and CIMT progression in these patients (n = 101). As a result, 22 DGs among the 26 identified DGs and total DG (β = 0.398, p < 0.001) were significantly associated with the annual change in CIMT. Furthermore, plasma DG levels improved the predictive ability for CIMT progression, with an adjusted R-squared increase of 0.105 [95% confidence interval: 0.010, 0.232] in the models. Plasma DGs are associated with CIMT progression in patients with type 2 diabetes. Measurement of total plasma DG levels may be beneficial in assessing the risk of atherosclerosis progression. Full article

Ocular diseases can significantly affect vision and overall quality of life, with diagnosis often being time-consuming and dependent on expert interpretation. While previous computer-aided diagnostic systems have focused primarily on medical imaging, this paper proposes VisionTrack, a multi-modal AI system for predicting multiple retinal diseases, including Diabetic Retinopathy (DR), Age-related Macular Degeneration (AMD), Diabetic Macular Edema (DME), drusen, Central Serous Retinopathy (CSR), and Macular Hole (MH), as well as normal cases. The proposed framework integrates a Convolutional Neural Network (CNN) for image-based feature extraction, a Graph Neural Network (GNN) to model complex relationships among clinical risk factors, and a Large Language Model (LLM) to process patient medical reports. By leveraging diverse data sources, VisionTrack improves prediction accuracy and offers a more comprehensive assessment of retinal health. Experimental results demonstrate the effectiveness of this hybrid system, highlighting its potential for early detection, risk assessment, and personalized ophthalmic care. Experiments were conducted using two publicly available datasets, RetinalOCT and RFMID, which provide diverse retinal imaging modalities: OCT images and fundus images, respectively. The proposed multi-modal AI system demonstrated strong performance in multi-label disease prediction. On the RetinalOCT dataset, the model achieved an accuracy of 0.980, F1-score of 0.979, recall of 0.978, and precision of 0.979. Similarly, on the RFMID dataset, it reached an accuracy of 0.989, F1-score of 0.881, recall of 0.866, and precision of 0.897. These results confirm the robustness, reliability, and generalization capability of the proposed approach across different imaging modalities. Full article

Biosensors play a central role in the early detection of abnormal glucose levels in individuals with diabetes; therefore, the development of less invasive systems is essential. Herein, a 3D-printed colorimetric biosensor combining microneedles and chitosan nanoparticles was developed for glucose detection in sweat using machine learning. Briefly, hollow 3D-printed polylactic acid microneedles were constructed and loaded with chitosan nanoparticles encapsulating glucose oxidase, horseradish peroxidase, and the chromogenic substrate 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid), resulting in the formation of the chitosan nanoparticle−microneedle patches. Glucose detection was performed colorimetrically by first incubating the chitosan nanoparticle−microneedle patches with glucose samples of varying concentrations and then by using photographs of the top side of each microneedle and a color recognition application on a smartphone. The Random Sample Consensus algorithm was used to train a simple linear regression model to predict glucose concentrations in unknown samples. The developed biosensor system exhibited a good linear response range toward glucose (0.025−0.375 mM), a low limit of detection (0.023 mM), a limit of quantification (0.078 mM), high specificity, and recovery rates ranging between 86–112%. Lastly, the biosensor was applied to glucose detection in spiked artificial sweat samples, confirming the potential of the proposed methodology for glucose detection in real samples. Full article

Pancreatic ductal adenocarcinoma (PDAC) remains among the most lethal malignancies, with a five-year survival rate below 12%, largely attributable to its asymptomatic onset, late-stage diagnosis, and limited curative treatment options. Although PDAC accounts for approximately 3% of all cancers, it is projected to become the second leading cause of cancer-related mortality in the United States by 2030. A major contributor to its dismal prognosis is the lack of validated early detection strategies for asymptomatic individuals. In this review, we present a comprehensive synthesis of current advances in the early detection of PDAC, with a focus on the identification of high-risk populations, novel biomarker platforms, advanced imaging modalities, and artificial intelligence (AI)-driven tools. We highlight high-risk groups—such as those with new-onset diabetes after age 50, pancreatic steatosis, chronic pancreatitis, cystic precursor lesions, and hereditary cancer syndromes—as priority populations for targeted surveillance. Novel biomarker panels, including circulating tumor DNA (ctDNA), miRNAs, and exosomes, have demonstrated improved diagnostic accuracy in early-stage disease. Recent developments in imaging, such as multiparametric MRI, contrast-enhanced endoscopic ultrasound, and molecular imaging, offer improved sensitivity in detecting small or precursor lesions. AI-enhanced radiomics and machine learning models applied to prediagnostic CT scans and electronic health records are emerging as valuable tools for risk prediction prior to clinical presentation. We further refine the Define–Enrich–Find (DEF) framework to propose a clinically actionable strategy that integrates these innovations. Collectively, these advances pave the way for personalized, multimodal surveillance strategies with the potential to improve outcomes in this historically challenging malignancy. Full article

The global healthcare system faces significant challenges posed by diabetes and its complications, highlighting the need for innovative strategies to improve early diagnosis and treatment. Machine learning models help in the early detection of diseases and recommendations for taking safety measures and treating the disease. A comparative analysis of existing machine learning (ML) models is necessary to identify the most suitable model while uniformly fixing the model parameters. Assessing risk based on biomarker measurement and computing overall risk is important for accurate prediction. Early prediction of complications that may arise, based on the risk of diabetes and biomarkers, using machine learning models, is key to helping patients. In this paper, a comparative model is presented to evaluate ML models based on common model characteristics. Additionally, a risk assessment model and a prediction model are presented to help predict the occurrence of complications. Random Forest (RF) is the best model for predicting the occurrence of Type 2 Diabetes (T2D) based on biomarker input. It has also been shown that the prediction of diabetes complications using neural networks is highly accurate, reaching a level of 98%. Full article

Over the past century, the understanding of type 1 diabetes mellitus (T1DM) has evolved significantly, transitioning from a fatal metabolic disorder to a well-characterized autoimmune disease. This review explores the historical developments and scientific milestones that have reshaped the perception of T1DM, highlighting key discoveries and shifts in medical paradigms. Methods: A comprehensive narrative review was conducted, examining literature spanning from ancient medical texts to contemporary research up to 2024. Emphasis was placed on pivotal moments such as the discovery of insulin in 1921, the recognition of autoimmune mechanisms in the 1970s, and recent advancements in immunotherapy. Results: The reclassification of T1DM as an autoimmune disease was supported from multiple lines of evidences including the presence of islet cell autoantibodies, the identification of lymphocytic infiltration in pancreatic islets, and the associations of the disease with certain HLA class II alleles. The development of animal models and large-scale cohort studies facilitated the establishment of disease staging and risk prediction models. Notably, the approval of immunotherapies like teplizumab underscores the translational impact of these scientific insights. Conclusions: The historical trajectory of T1DM exemplifies the dynamic nature of medical knowledge and the interplay between clinical observations and scientific research. Recognizing these developments enhances our comprehension of disease mechanisms and informs current approaches to diagnosis and treatment. Full article

Peripheral artery disease (PAD) is associated with an elevated risk of major adverse cardiovascular events (MACE). Despite this, few reliable biomarkers exist to identify patients at heightened risk of MACE. Growth differentiation factor 15 (GDF15), a stress-responsive cytokine implicated in inflammation, atherosclerosis, and thrombosis, has been broadly studied in cardiovascular disease but remains underexplored in PAD. This study aimed to evaluate the prognostic utility of GDF15 for predicting 2-year MACE in PAD patients using explainable statistical and machine learning approaches. We conducted a prospective analysis of 1192 individuals (454 with PAD and 738 without PAD). At study entry, patient plasma GDF15 concentrations were measured using a validated multiplex immunoassay. The cohort was followed for two years to monitor the occurrence of MACE, defined as stroke, myocardial infarction, or death. Baseline GDF15 levels were compared between PAD and non-PAD participants using the Mann–Whitney U test. A machine learning model based on extreme gradient boosting (XGBoost) was trained to predict 2-year MACE using 10-fold cross-validation, incorporating GDF15 and clinical variables including age, sex, comorbidities (hypertension, diabetes, dyslipidemia, congestive heart failure, coronary artery disease, and previous stroke or transient ischemic attack), smoking history, and cardioprotective medication use. The model’s primary evaluation metric was the F1 score, a validated measurement of the harmonic mean of the precision and recall values of the prediction model. Secondary model performance metrics included precision, recall, positive likelihood ratio (LR+), and negative likelihood ratio (LR-). A prediction probability histogram and Shapley additive explanations (SHAP) analysis were used to assess model discrimination and interpretability. The mean participant age was 70 ± SD 11 years, with 32% (n = 386) female representation. Median plasma GDF15 levels were significantly higher in PAD patients compared to the levels in non-PAD patients (1.29 [IQR 0.77–2.22] vs. 0.99 [IQR 0.61–1.63] pg/mL; p < 0.001). During the 2-year follow-up period, 219 individuals (18.4%) experienced MACE. The XGBoost model demonstrated strong predictive performance for 2-year MACE (F1 score = 0.83; precision = 82.0%; recall = 83.7%; LR+ = 1.88; LR− = 0.83). The prediction histogram revealed distinct stratification between those who did vs. did not experience 2-year MACE. SHAP analysis identified GDF15 as the most influential predictive feature, surpassing traditional clinical predictors such as age, cardiovascular history, and smoking status. This study highlights GDF15 as a strong prognostic biomarker for 2-year MACE in patients with PAD. When combined with clinical variables in an interpretable machine learning model, GDF15 supports the early identification of patients at high risk for systemic cardiovascular events, facilitating personalized treatment strategies including multidisciplinary specialist referrals and aggressive cardiovascular risk reduction therapy. This biomarker-guided approach offers a promising pathway for improving cardiovascular outcomes in the PAD population through precision risk stratification. Full article

Background/Objectives: A healthy diet helps prevent noncommunicable diseases, and dairy is an essential part of this diet. Multiple meta-analyses have shown an inverse association between yogurt intake and type 2 diabetes (T2D). This study aimed to develop a simulation model and evaluate the medical and economic effects of increased yogurt intake on T2D. Methods: It predicted the T2D incidence rate, T2D mortality rate, and national healthcare expenditures (NHE) over 10 years using a Markov model for the Japanese population aged 40–79 years. Results: By increasing yogurt intake to 160 g/day or 80 g/day, the incidence rate of T2D decreased by 16.1% or 5.9%, the T2D-related mortality rate decreased by 1.6% or 0.6%, and the NHE was predicted to decrease by 2.4% and 0.9%, respectively. Conclusions: Increasing yogurt intake may be an effective strategy to prevent T2D and reduce NHE. Full article

Background and Objectives: Bell’s palsy, characterized by acute idiopathic facial nerve paralysis, exhibits variable recovery outcomes influenced by treatment timing, modality, and patient comorbidities. This study aimed to compare the effectiveness of corticosteroid-based treatment (Ear, Nose, and Throat [ENT]), nerve blocks/physical therapy (Pain Medicine), and acupuncture/herbal medicine (Traditional Korean Medicine [KM]) and identify predictors of recovery and recurrence. This retrospective cohort study leverages South Korea’s pluralistic healthcare system, where patients choose specialties, to provide novel insights into departmental treatment outcomes. Materials and Methods: We analyzed 600 patients treated within 72 h of Bell’s palsy onset (2010–2024) at Wonkwang University Hospital, South Korea, using propensity score matching (PSM) (1:1:1) for age, sex, comorbidities, and initial House–Brackmann (HB) grade. The primary outcome was complete recovery (HB grade I) at 6 months; secondary outcomes included recovery time, recurrence, complications, and patient satisfaction. Multivariate logistic regression identified predictors. Results: The ENT group achieved the highest complete recovery rate (87.5%, phi = 0.18) versus Pain Medicine (74.0%) and KM (69.5%) (p < 0.001), with the shortest recovery time (4 weeks, Cohen’s d = 0.65 vs. KM). Synkinesis was lowest in the ENT group (6.0%). ENT treatment (OR: 1.75; 95% CI: 1.29–2.37) and early corticosteroid application (OR: 1.95; 95% CI: 1.42–2.68) predicted recovery. Hypertension (OR: 4.40), hyperlipidemia (OR: 8.20), and diabetes (OR: 1.40) increased recurrence risk. Subgroup analyses showed that ENT treatment was most effective for severe cases (HB grade IV: 90% recovery vs. 65% in KM, p < 0.01). Conclusions: Corticosteroid-based treatment (ENT) yielded superior recovery outcomes. Comorbidity management is critical for recurrence prevention. Early ENT referral and integrated care models are recommended to optimize outcomes in diverse healthcare settings. Full article

Background and Objectives: Cardiovascular (CV) diseases account for about 32% of deaths in women, with differing risk factors between women and men. Our study aimed to compare sex-related risk factors and comorbidities in patients at very high CV risk. Materials and Methods: We consecutively enrolled adult patients hospitalized for myocardial infarction or unstable angina at a tertiary referral center in western Romania between October 2016 and June 2017. A total of 299 adults underwent clinical and biochemical evaluations between 6 months and 2 years after their coronary event. We assessed patients’ specific characteristics, comorbidities, and risk factors. Results: Women made up only a quarter of the survey participants (74 women, 24.7%) and were generally older (63.32 ± 9.3 vs. 60.51 ± 9.3, p = 0.02) and more obese (31.20 ± 6.0 vs. 29.48 ± 4.9, p = 0.02). There were no significant differences in the prevalence of hypertension, diabetes, dyslipidemia, chronic kidney disease, or peripheral artery disease, though women had slightly higher rates for most comorbidities. Regarding smoking habits, both groups had high percentages of current and former smokers, with women being significantly less likely to smoke (20.9% vs. 44.6%, p = 0.003). Multivariable logistic regression adjusting for age, BMI, smoking status, diabetes, and eGFR revealed that sex was not a statistically significant independent predictor for myocardial infarction, PCI, or CABG. Conclusions: We observed that women with previous coronary events had a worse risk factor profile, while there were no significant sex differences in overall comorbidities. Risk factor control should be based on sex-specific prediction models. Full article