Browse Articles
Discover research articles across all indexed journals
Diabetic retinopathy as the primary predictor of mild cognitive impairment in type 2 diabetes: Insights from machine learning models
Mild cognitive impairment (MCI) is a significant and increasingly recognized problem in individuals with type 2 diabetes mellitus (T2DM). This study aims to develop a machine-learning model to predict MCI in patients with T2DMThe dataset was obtained from a prospective cohort conducted at the Sheikh Khalifa Ibn Zaid Hospital, Casablanca, Morocco, and was randomly split into three parts. Machine learning models were trained using the training dataset, and their performance was assessed on the test dataset. The unseen data was reserved for final validation. Subsequently, recursive feature elimination was applied with the top two algorithms to identify and retain the most impactful features for predicting MCI. Then, we retrained the models using the selected variables. Finally, the variables most contributing to the prediction of MCI were represented in a Shapley Additive Explanations SHAPE value plot to better understand their contribution and their ranking in MCI prediction.The dataset included 100 patients. Extra Trees classifier was the best-performing model for mild cognitive impairment (Accuracy: 0·9310/AUC: 0·9667/ Recall: 0·9333/ Precision: 0·9333/F1: 0·9333). The most contributing factors to MCI in patients with type 2 diabetes were, respectively, diabetic retinopathy, age, serum LDL cholesterol level, microalbuminuria, HbA1c, and, serum creatinine level. Our findings suggest avenues for early intervention that could prevent the progression of cognitive impairment among patients with type 2 diabetes.
Quantification of hounsfield unit difference in heterogeneous hematoma predicts poor functional outcomes in acute intracerebral hemorrhage
Enhanced photocatalytic degradation of penicillin G using magnetic silica doped ZnAl- layered double hydroxides
Edge percolation centrality: A new measure to quantify the influence of edges during percolation in networks
Numerous centrality measures exist to quantify the influence of edges within a network, with edge betweenness being one of the more well-known measures. However, such measures are inadequate in network percolation scenarios (e.g., the transmission of a disease over a transportation network of highways) as they fail to consider the changing percolation states of edges over time. This paper addresses this limitation by extending percolation centrality, a measure originally developed to evaluate the influence of vertices during a percolation process (i.e., a dynamic spread of a contagion) in the network, to the edge level. The proposed measure, edge percolation centrality, captures both the topological connectivity of the network as well as the percolation states of the edges. Although the algorithm’s observed complexity of O(|V|3.57) makes it computationally intensive, the utility of the proposed edge measure is evident in its application to both synthetic and real-world networks undergoing percolation processes.
Assessment of suitable sub-watersheds for sand dams and prioritization of artificial sandy aquifers for sustainable groundwater development
Novel concept of linguistic fractional fuzzy information for effective water filtration decision-making problem based on WASPAS method
Construction of an automated machine learning-based predictive model for postoperative pulmonary complications risk in non-small cell lung cancer patients undergoing thoracoscopic surgery
Objective To develop a predictive framework integrating machine learning and clinical parameters for postoperative pulmonary complications (PPCs) in non-small cell lung cancer (NSCLC) patients undergoing video-assisted thoracic surgery (VATS). Methods This retrospective study analyzed 286 NSCLC patients (2022–2024), incorporating 13 demographic, metabolic-inflammatory, and surgical variables. An Improved Blood-Sucking Leech Optimizer (IBSLO) enhanced via Cubic mapping and opposition-based learning was developed. Model performance was evaluated using AUC-ROC, F1-score, and decision curve analysis (DCA). SHAP interpretation identified key predictors. Results The IBSLO demonstrated significantly superior convergence performance versus original BSLO, ant lion optimizer (ALO), Harris hawks optimization (HHO), and whale optimization algorithm (WOA) across all 12 CEC2022 test functions. Subsequently, the IBSLO-optimized automated machine learning (AutoML) model achieved ROC-AUC/PR-AUC values of 0.9038/0.8091 (training set) and 0.8775/0.8175 (testing set), significantly outperforming four baseline models: logistic regression (LR), support vector machine (SVM), XGBoost, and LightGBM. SHAP interpretability identified six key predictors: preoperative leukocyte count, body mass index (BMI), surgical approach, age, intraoperative blood loss, and C-reactive protein (CRP). Decision curve analysis demonstrated significantly higher net clinical benefit of the AutoML model compared to conventional methods across expanded threshold probability ranges (training set: 8–99%; testing set: 3–80%). Conclusion This study establishes an interpretable machine learning framework that improves preoperative risk stratification for NSCLC patients, offering actionable guidance for thoracic oncology practice.
Automated detection and classification of cervical and anal squamous cancer precursors using deep learning and multidevice colposcopy
Diagnostic applicability of the circulating growth factors in preeclampsia
Abstract This study aimed to compare the levels of various growth factors in maternal serum and umbilical vein serum between individuals with preeclampsia (PE) and those with normotensive pregnancies. One hundred eight pregnant women were enrolled in this prospective study. Blood samples were taken at third trimester. 67 serum samples of preeclampsia and 41 normotensive mothers were collected. Umbilical vein serum samples of 27 preeclampsia and 38 normotensive pregnancies were collected. Serum levels of BDNF, EGF, FGF-2, HGF, LIF, NGF-beta, PDGF-BB, PlGF-1, SCF, and vascular endothelial growth factor (VEGF)-A and D were measured. EGF, HGF, LIF, and SCF levels were significantly higher in PE compared to normotensive group after adjustment for gestational age. There were no differences in growth factors in umbilical vein serum between two groups. EGF, NGF-beta and SCF were associated with the elevation of systolic and diastolic blood pressure. SCF showed a positive correlation with ALT and creatinine and a decreasing trend as gestational age advances. The ROC curve revealed cutoff value of 36.6 pg/mL in EGF, 23.5 in HGF, 2.78 in LIF and 2.95 in SCF (p < 0.001, p = 0.042, p < 0.001, p < 0.001). Circulating growth factors in PE may be responsible for development or consequence associated with the pathogenesis in PE especially elevated EGF, HGF, LIF and SCF.
Gene electrotransfer with flow-through microchannel and lower alternating voltage generated induced pluripotent cells from human lymphoblastoid cell lines
Induced pluripotent stem cells (iPSCs) are useful for studying genetic and rare diseases and can be generated by reprogramming immortalized lymphoblastoid cell lines (LCLs) stored in global repositories with detailed genotype and phenotype data. Traditional bulk-type electroporators are commonly used for gene electrotransfer in reprogramming, but they have major drawbacks, including high costs associated with electric pulse generators and the requirement for fixed volumes for costly reprogramming factors. These limitations hinder cost-effective and scalable iPSC generation, particularly when working with large numbers of LCLs with diverse genotypes. We aimed to develop a flow-through-type electroporator utilizing microchannels for the generation of iPSCs from LCLs, to reduce the costs associated with traditional bulk-type electroporators and enable parallel processing for LCLs with various genotypes. We applied a continuous wave of biphasic alternating voltage (~10 V one-sided amplitude) to micro-scaled electrodes within the microchannel to develop a flow-through electroporator. Numerical simulations were conducted to assess the electric field distribution and its applicability to pore formation in the plasma membrane. To optimize electroporation and flow conditions, we used plasmid pCXLE-EGFP (encoding Green Fluorescent Protein, GFP) for gene electrotransfer to LCLs. Reprogramming factors (pCXLE-hSK, pCXLE-hOCT3/4-shp53-F, pCXLE-hUL) were also delivered to the cells via the same system. The flow-through electroporator achieved 31% transfection efficiency with 78% cell viability, 2 d post-electroporation. In each condition, only 3 µL of cell suspension was used with 10 7 cells/mL of cells and 500 ng/µL plasmid vector. A reprogramming efficiency of 0.048% was obtained, which is comparable to that achieved using bulk-type electroporators. This developed flow-through electroporator with microchannel technology offers significant advantages over traditional methods, including the potential to reduce costs and the ability to process small volumes of cell suspension, making it suitable for parallel processing of LCLs with diverse genotypes. The system provides a promising approach for scalable and potentially cost-effective iPSC generation.
Association between enterotypes of the gut microbiota and features of ischemic stroke
Secure and fault tolerant cloud based framework for medical image storage and retrieval in a distributed environment
Postural control and trunk muscle activation in transtibial amputees: A pilot electromyographic study
Transtibial amputation affects postural stability, requiring trunk muscle analysis to support balance and rehabilitation. This pilot study aimed to compare trunk muscle activations during postural sway and limits of stability in unilateral transtibial amputees with healthy controls and investigate the relationship between postural control and trunk muscle activations in transtibial amputees. Accordingly, it was hypothesised that transtibial amputees would exhibit altered postural control and trunk muscle activation, both compared to controls and between limbs. This preliminary observational cross-sectional study included a transtibial amputee group (n = 10) and a healthy control group (n = 10). Static and dynamic balance were assessed using a Bertec force platform. Trunk muscle activation was measured bilaterally using the Delsys wireless surface electromyography system while balance was assessed. Subsequently, group differences in postural control and trunk muscle activation were analyzed. Transtibial amputee group showed significantly greater limits of stability distance on the amputated side compared to controls’ non-dominant side (p < 0.05). Additionally, transtibial amputee group exhibited greater lateral postural sway under compliant surface conditions compared to healthy controls (p < 0.05). During limits of stability assesment, external oblique activation was higher on the intact side of transtibial amputee group compared to controls’ dominant side and higher than on the amputated side (p < 0.05). Multifidus activation on the amputated side during compliant surface with eyes open was greater than the intact side (p < 0.05). Longissimus dorsi activation on the intact side in amputee group exceeded controls’ dominant side under all conditions (p < 0.05). Moreover, longissimus dorsi activation on the amputated side was significantly higher than on the controls’ nondominant side during compliant surface with eyes closed (p < 0.05). Preliminary findings highlight disrupted postural control and trunk muscle activation in transtibial amputees, indicating the need for targeted rehabilitation and larger studies.
Shoreline change detection along the eastern coast of Ghana: leveraging GIS and advanced remote sensing techniques
Anti-inflammatory activity of Leonotis nepetifolia leaf essential oil in LPS-stimulated RAW 264.7 cells and its molecular mechanism of action
Impacts of resistance training combined with vibration training on the IGF-1/PI3K/AKT/FOXO3 axis and clinical outcomes in patients with sarcopenia: A protocol for a randomized controlled trial
Background Sarcopenia is an age-associated disorder characterized by a progressive decline in skeletal muscle mass, strength, and physical function. The condition is linked to low levels of anabolic hormones such as insulin-like growth factor 1 (IGF-1), with its downstream phosphatidylinositol 3 kinase (PI3K)/ protein kinase B (AKT)/ forkhead box protein O3 (FOXO3) signaling pathway. There is growing evidence that resistance training (RT) or vibration training (VT) could improve physical functioning in individuals with sarcopenia. However, the related physiological influence of exercise on sarcopenia remains elusive. Method This prospective randomized controlled trial will be conducted among 96 participants, aged between 65 and 80 years. In participants, sarcopenia diagnosis will be confirmed based on the Asian Working Group for Sarcopenia criteria, and participants will be randomized into either control, RT, VT, or RVT (combined RT and VT) groups. The intervention will last 12 weeks, with assessments performed at baseline, 12 weeks (after intervention), and 24 weeks (follow-up). The primary outcomes will include skeletal muscle mass, handgrip strength, and gait speed. Secondary outcomes comprise IGF-1 concentrations, PI3K/AKT and FOXO3 protein activity, quality of life, and timed-up-and-go test performance assessments. Discussion This clinical study aims to elucidate the potential modulation of molecular mechanisms in vivo for combined RT and VT in sarcopenia patients and to identify the effects of the intervention on physical function. Trial registration ChiCTR, ChiCTR2400083643. Registered on April 29, 2024.
ANN-based thermal analysis of 3D MHD hybrid nanofluid flow over a shrinking sheet via LMA
Identification of camouflage military individuals with deep learning approaches DFAN and SINETV2
How do digital platform capabilities affect service innovation in service manufacturing firms? The role of knowledge creation and environmental uncertainty
The recent wave of scientific and technological revolutions, coupled with industrial advancements, has catalyzed the rise of service-oriented manufacturing enterprises in China. Despite their notable achievements, these enterprises grapple with a “service dilemma,” marked by a misalignment between input and output and ineffective service innovation. Digital platform capabilities emerge as a pivotal factor influencing service innovation. However, research has not explored their underlying mechanisms in depth. Therefore, this study investigates the impact of digital platform capabilities on service innovation within service-oriented manufacturing enterprises, examining the mediating role of knowledge creation and the moderating role of environmental uncertainty. Data were collected via structured questionnaires from middle and senior managers of service-oriented manufacturing enterprises, and hierarchical regression analysis was used to test the hypotheses. The findings reveal that digital platform capabilities significantly enhance service innovation. Moreover, knowledge creation mediates the relationship between digital platform capabilities and service innovation, while environmental uncertainty negatively moderates the link between knowledge creation and service innovation. Based on these insights, this paper proposes strategic recommendations to bolster service innovation by emphasizing digital platform capabilities, knowledge creation, and environmental uncertainty. These strategies aim to support the sustainable growth of service-oriented manufacturing enterprises in the rapidly evolving digital landscape.