Browse Articles

Development and validation of a risk prediction algorithm for high-risk populations combining genetic and conventional risk factors of cardiovascular disease

Aim To develop a model for cardiovascular disease (CVD) risk, combining polygenic risk score (PRS) with traditional risk factors while assessing the added value of PRS in two cohorts of biobank participants. Methods Data of 128 209 participants from the Estonian Biobank recruited between 2002–2017 and 2018–2022 without prevalent cardiovascular disease, was included. Hazard ratios (HR) for polygenic risk versus conventional risk factors were estimated with Cox proportional hazards models, cumulative incidence was assessed with Aalen-Johansen curves. Predictive performance was tested using a split-sample approach and competing risk modelling. Age at CVD event served as the outcome, and the impact of the PRS was evaluated by age group (25–59 vs. 60+), sex, and recruitment period, using HRs, Harrell’s C-index, and net reclassification indices (NRI). Results The estimated HR per one standard deviation (SD) of PRS ranged from 1.1, 95% CI 1.06–1.15 (age 60 + , earlier cohort) to 1.36, 95% CI 1.24–1.49 (men 25–59, later cohort). Adding PRS to the conventional risk factors in the age group 25–59 increased the C-statistic by 0.028 (p < 0.0001) for men. In the age group 60 + , the increase was 0.016 (p = 0.0002) across all. In the independent validation set, the continuous NRI was 19.1% (95% CI 13.3%–24.9%) in the 25–59 group and 13.9% (95% CI 8.1%–19.6%) in the 60 + group. Conclusions In a high-risk population, PRS is a strong independent risk factor for CVD and should be considered in routine risk assessment, starting at a relatively young age.

Multi-strategy dung beetle optimization for robust indoor object detection and tracking for visually impaired people with hybrid deep learning networks

Research on grape leaf classification based on optimized densenet201 model

In the realm of plant classification, the classification of grape leaf varieties has long presented a complex challenge. Aiming to enhance the accuracy and generalization ability of grape leaf variety classification, this study proposes a novel approach that employs an optimized Densenet201 model for grape leaf classification. Initially, grape leaf images from five distinct varieties were meticulously collected to construct a comprehensive grape leaf dataset. To augment the diversity of the dataset, the parameters of data augmentation were delicately adjusted, with an increase in the rotation range, translation range, and so on. Subsequently, BatchNormalization and GlobalAveragePooling2D layers were incorporated to achieve feature normalization and pooling. Simultaneously, the parameters of the Dropout layer were optimized to effectively mitigate the issue of overfitting. Additionally, the number of neurons and layers in the Dense layer were varied to explore diverse network structures and pursue superior performance. Moreover, the parameters of the Adam optimizer were meticulously tuned to attain the optimal performance, and the model’s performance was further enhanced by extracting image features. The experimental results demonstrate that, in comparison with the densenet121, densenet169, resnet50, and densenet201 models, the optimized Densenet201 model showcases outstanding performance in grape leaf variety classification, remarkably improving the classification accuracy and generalization ability. This research provides a more efficient method for grape leaf variety classification.

Investigation of mechanical properties of 316 l steel samples at slm process with ultrasonic influence

Development and validation of a dual language needs assessment tool for people living with colorectal cancer (NeAT-CC)

Background Needs assessment tools may guide optimization of clinical services to be more patient-centered. As needs of patients living with and beyond colorectal cancer (CRC) may also be influenced by socio-cultural backgrounds and healthcare ecosystems, we developed and validated a needs assessment questionnaire for CRC in a multi-ethnic, low-and middle-income setting. Methods The study methodology was guided by the COSMIN checklist. Items generation was based on findings from independent qualitative inquiries with patients, input from cancer stakeholders, and literature review. Following translation into Malay language, content and face validation were undertaken. The tool was administered to 300 individuals living with and beyond CRC. Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) were performed. Criterion validity was assessed using EORTC QLQ-C30 and QLQ-CR29 questionnaires. Results The 48-item bilingual needs assessment tool for colorectal cancer (NeAT-CC) encompassed six domains of needs, namely (i) diagnosis, (ii) psychosocial and information, (iii) healthcare, (iv) practical and living with cancer, (v) financial and (vi) employment. Cronbach’s alpha was above 0.70 for all domains, indicating good internal consistency. CFA also demonstrated acceptable convergent and divergent validity with composite reliability >0.70 and Heterotrait–Monotrait index <0.90 for all constructs. Criterion validity was established given the significant correlation with quality of life. The NeAT-CC was easily understandable, took 15–20 minutes for completion and may be self-administered. Conclusions Utilization of NeAT-CC may facilitate optimization of supportive and survivorship care services following CRC in local settings. The tool has wider potential for adaptation in other multi-ethnic and/or low and middle-income settings.

Collaborative optimization of layout and cutting scheduling for large-scale customized metal structural parts

Do lifestyle factors affect patient reported clinical outcomes after total knee replacement surgery? A feasibility cohort study (PRO-Knee)

Aims To evaluate the feasibility of a substantive cohort study to determine whether modifiable lifestyle factors, including smoking, physical inactivity, alcohol consumption and being overweight, affect patient-reported clinical outcomes after total knee replacement surgery. Methods Adults awaiting total knee replacement surgery were recruited pre-operatively and completed self-reported questionnaires at baseline and 3- and 6-months post-surgery. Feasibility outcomes, including recruitment, retention and response rate of the primary outcome questionnaire were analysed descriptively. Results 40 participants were recruited from 183 eligible patients (22%). 87.5% (35/40) participants returned questionnaires at 6-months. 85% (34/40) of participants were overweight (BMI > 24.9), 25% (10/40) drank alcohol (AUDIT-C > 4), 5% (2/40) smoked tobacco and 67.5% (27/40) were physically inactive (GPPAQ classification of ‘moderately inactive’, or ‘inactive’). Conclusion Modifiable lifestyle factors including smoking, alcohol use, physical inactivity and being overweight are highly prevalent in patients waiting for total knee replacement. Based on this study, a future substantive cohort study investigating the effect of lifestyle factors on clinical outcomes post total knee replacement in the UK NHS is feasible.

Multidimensional factors of health-related quality of life in parkinson’s disease using ensemble learning and network analysis

Optimizing network bandwidth slicing identification: NADAM-enhanced CNN and VAE data preprocessing for enhanced interpretability

Communication networks of the future will rely heavily on network slicing (NS), a technology that enables the creation of distinct virtual networks within a shared physical infrastructure. This capability is critical for meeting the diverse quality of service (QoS) requirements of various applications, from ultra-reliable low-latency communications to massive IoT deployments. To achieve efficient network slicing, intelligent algorithms are essential for optimizing network resources and ensuring QoS. Artificial Intelligence (AI) models, particularly deep learning techniques, have emerged as powerful tools for automating and enhancing network slicing processes. These models are increasingly applied in next-generation mobile and wireless networks, including 5G, IoT infrastructure, and software-defined networking (SDN), to allocate resources and manage network slices dynamically. In this paper, we propose an Interpretable Network Bandwidth Slicing Identification (INBSI) system that leverages a modified Convolutional Neural Network (CNN) architecture with Nesterov-accelerated Adaptive Moment Estimation (NADAM) optimization. Additionally, we use a Variational Autoencoder (VAE) for preprocessing initial data, along with reconstructed data for data validity assessment. The model we propose outperforms other alternatives and reaches an accuracy peak of (84%) in the system environment. A range of accuracy was achieved by (k-nearest neighbors algorithm) KNN (76%), Random Forest (69%), BaggingClassifier (70%), and Gaussian Naive Bayes (GaussianNB) (55%). The accuracy of additional methods varies, including Decision Trees, AdaBoost, Deep Neural Forest (DNF), and Multilayer Perceptrons (MLPs). We utilize two eXplainable Artificial Intelligence (XAI) approaches, Shapley Additive Explanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME), to provide insight into the impact of certain input characteristics on the network slicing process. Our work highlights the potential of AI-driven solutions in network slicing, offering insights for operators to optimize resource allocation and enhance future network management.

Forelimb biomechanics in the derived therizinosaur Nothronychus and its relation to the origin of the avian wing

Efficacy of ceiling-mounted mosquito nets for malaria vector control in a Peruvian Amazon riverine community: A stepped-wedge cluster randomized trial

Malaria remains a major global health concern, especially in tropical regions such as the Peruvian Amazon. Installing ceiling-mounted mosquito nets in houses has been proposed as a strategy to reduce mosquito–human contact and, in turn, lower malaria transmission. We aimed to assess the effectiveness of ceiling-mounted mosquito nets in reducing Anopheles mosquito density in a high-transmission Amazonian setting. We conducted a stepped-wedge cluster-randomized trial from March to December 2024 in the Llanchama community of the San Juan Bautista district, Loreto, Peru. A total of 69 households were randomized into three clusters, receiving the intervention at staggered three-month intervals. We measured entomological indices using human landing catches (HLC) and other standardized methods, focusing on Anopheles mosquito counts, bites per person per night (BPN), and bites per person per hour (BPH), both indoors and peridomestic, derived from the same HLC sessions, using cumulative mosquito captures over 12-hour periods to compute biting indices. Houses were retrofitted with mosquito nets across ceilings and other open structural areas, creating a barrier to prevent mosquito entry and lower exposure. Analysis followed a stepped-wedge mixed-effects negative binomial model adjusting for clustering and time trends. Our per-protocol analysis shows that, compared to non-renovated households, the renovated households experienced a 55% reduction in indoor Anopheles counts (95% CI: 33%–74%; p = 0.004), a 60% decrease in indoor BPN (95% CI: 27%–78%; p = 0.003), and a 61% reduction in indoor BPH (95% CI: 15%–83%; p = 0.018). Peridomestic mosquito counts, BPN, and BPH did not differ significantly between renovated and non-renovated households. Our study provides evidence that installing mosquito nets across household ceilings markedly reduces indoor Anopheles presence and biting rates, suggesting that this structural modification could be a promising strategy for lowering malaria risk in riverine communities.

Improving phishing email detection performance through deep learning with adaptive optimization

Impact of early tracheostomy on resource utilization and patient outcomes in trauma ICU patients: A retrospective cohort study from southern India

Introduction A tracheostomy is an important intervention for trauma patients referred to intensive care units (ICUs). Trauma patients often require prolonged intubation; timing of tracheostomy remains debated.The purpose of this study is to determine the impact of early tracheostomy on critical metrics such as mechanical ventilation duration, ICU length of stay (LOS) and ventilator acquired pneumonia (VAP) in trauma patients in ICU settings. Methods We conducted a retrospective cohort study of 383 trauma patients who underwent tracheostomy in a tertiary teaching hospital ICU (January 2018–December 2022). Inclusion: trauma patients with temporary tracheostomy; Exclusion: permanent tracheostomies. Early tracheostomy (ET) was defined as ≤7 days of mechanical ventilation, late (LT) as >7 days. The dataset includes demographic information, Acute physiology and chronic health evaluation II score, Simplified acute physiology score II, Glasgow coma scale score, Injury severity Score, type and cause of injuries, ICU outcomes, length of stay and rates of ventilator-associated pneumonia (VAP). Data were analyzed using Mann–Whitney U and Chi-square tests; significance at p < 0.05.. The study involved a comparison of the duration of mechanical ventilation, ICU LOS, VAP rates and extubation trials between patients who underwent ET and LT. Results Of the 804 patients who underwent tracheostomies from January 2018 to December 2022, 383 were trauma patients and were included in the study. There were no significant differences between the two groups in terms of age, sex, Acute physiology and chronic health evaluation II score, Simplified acute physiology score II and Injury severity score. The incidence of VAP was lower in the ET cohort (15.9%) than in the LT cohort (47.4%). The percentage of extubation trials was found to be higher in the LT cohort (43.1%) than in the ET cohort (9.3%), resulting in prolonged ICU LOS. Patients with an ET had a significantly shorter ICU LOS median of 15 days (IQR 13,17) and a mechanical ventilation median of 13 days (IQR 11,14) than LT patients who had an ICU LOS median of 33 days (IQR 30,36) and a mechanical ventilation median of 31 days (IQR 27,33) respectively. Conclusion Implementing an early tracheostomy protocol for trauma patients in the ICU is associated with a decreased incidence of VAP, shorter duration of mechanical ventilation and shorter ICU LOS while maintaining consistent ICU and hospital outcomes. The adoption of a standardized approach to perform early tracheostomy helps in improving resource utilization and patient outcomes in trauma patients.

Interface-induced collective phase transition in VO2-based bilayers studied by layer selective spectroscopy

Abstract We investigated the origin of collective electronic phase transitions induced at the heterointerface between monoclinic insulating VO 2 and rutile metallic electron-doped VO 2 layers using in situ soft X-ray photoemission spectroscopy (PES) and X-ray absorption spectroscopy (XAS) on nanoscale VO 2 /V 0.99 W 0.01 O 2 (001) R bilayers. Thanks to the surface sensitivity of PES and XAS, we determined the changes in the electronic structure and V–V dimerization in each constituent layer separately. The layer selective observation of the electronic and crystal structures in the upper VO 2 layer of the bilayer indicates that the monoclinic insulating phase VO 2 layer undergoes a transition to the rutile metallic phase by forming the heterointerface. Detailed temperature-dependent measurements reveal that the rutile metallic phase VO 2 undergoes a transition to the monoclinic insulating phase with a decrease in temperature, as in the case of a VO 2 single-layer film. Furthermore, during the temperature-induced phase transition in the VO 2 layer, the spectra are well described by an in-plane phase separation of the rutile metallic and monoclinic insulating phases. These results suggest that the interface-induced transition from the monoclinic insulating to the rutile metallic phase in the VO 2 layer of bilayers occurs as a collective phase transition derived from the static energy balance between the interfacial energy and the bulk free energies of the constituent layers.

Colorimetric detection of chloroperoxyl radical in reactive chlorine species solutions

Chlorous acid water (CAW) is a chlorine-based disinfectant approved as a food additive in Japan. CAW is synthesized by the reaction of chloric acid aqueous solution with hydrogen peroxide under acidic conditions. However, in this synthesis method, various ions — such as Na + from NaClO 3 and the conjugate base of the acid used — remain in the solution, hindering the selective detection and decreasing stability of chloroperoxyl radical (ClOO•), a potential key disinfectant species. In this study, we aimed to establish a colorimetric quantification method for ClOO•. We prepared a high-purity ClOO• solution (ClO 2_cx ) by cation exchange and its purity and stability were evaluated using electron spin resonance (ESR) spectroscopy and total chlorine concentration measurements. Furthermore, several colorimetric methods—including DPD, TMB, and DPPH assays—were examined to quantify ClOO•, and their sensitivity and selectivity were comparatively assessed. ClOO• was the sole detectable oxidant in the solution, with a lifetime exceeding 100 h, indicating exceptional stability under ambient conditions. Among the evaluated colorimetric methods, the DPD-based method was found to be suitable for quantifying ClOO•, showing a wide detection range and excellent linearity. This study represents the first report of a colorimetric quantification method for ClOO•. Our findings are expected to be useful for quantitatively discussing the biological efficacy of ClOO• and its reaction mechanisms.

Cloud-based DDoS detection using hybrid feature selection with deep reinforcement learning (DRL)

Development and validation of a bedside-available machine learning model to predict discrepancies between SaO₂ and SpO₂: Exploring factors related to the discrepancies

In critically ill patients, a discrepancy frequently exists between percutaneous oxygen saturation (SpO₂) and arterial blood oxygen saturation (SaO₂), which can lead to potential hypoxemia being overlooked. The aim of this study was to explore the factors related to the discrepancy and to develop an easy-to-use prediction model that uses readily available bedside information to predict the discrepancy and suggest the need for arterial blood gas measurement. This is a prognostic study that used eICU Collaborative Research Database from 2014 to 2015 for model development and MIMIC-IV data from 2008 to 2019 for model validation. To predict the outcome of SpO₂ exceeding SaO₂ by 3% or more, non-invasive, readily available bedside information (patient demographics, vital signs, vasopressor use, ventilator use) was used to develop prediction models with three machine learning methods (decision tree, logistic regression, XGBoost). To make the model accessible, the model was deployed as a web-based application. Additionally, the contribution of each variable was explored using partial dependence plots and SHAP values. From 4,781 admission records in eICU data, a total of 19,804 paired SpO₂ and SaO₂ measurements were used. Among three machine learning models, the XGBoost model demonstrated the best predictive performance with an AUROC of 0.73 and a calibration slope of 0.90. In the validation cohort of MIMIC-IV paired dataset, the performance was AUROC of 0.56. An exploratory model-updating step followed by temporal validation raised performance to AUROC of 0.70 with a calibration slope of 0.85. In both datasets, worse vital signs were associated with the discrepancy (e.g., low blood pressure, low temperature) between SpO₂ and SaO₂. Using non-invasive bedside data, a machine learning model was developed to predict SpO₂–SaO₂ discrepancy and identified vital signs as key contributors. These findings underscore the awareness for hidden hypoxemia and provide the basis of further study to accurately evaluate the actual SaO₂.

Text mining and machine learning reveal global determinants of food insecurity

Prevalence of opportunistic bacterial infections (tuberculosis and pneumonia) among people with HIV in Ethiopia: Systematic review and meta-analysis

Background Individuals with weakened immune systems, such as those living with Human Immunodeficiency Virus (HIV), are more vulnerable to opportunistic bacterial infections, which include tuberculosis and pneumonia. This systematic review and meta-analysis looked at the pooled prevalence of opportunistic bacterial infections among people living with HIV in different regions of Ethiopia. Methods By looking through open online databases, articles written in English were considered. Joanna Briggs Institute’s critical appraisal tool for prevalence study was used to check the quality of each article. Inverse variance (I 2 ), sensitivity analysis, funnel plot, and Egger’s regression tests were used to check heterogeneity and publication bias. Because of a high heterogeneity, a random-effects model was used to estimate the pooled prevalence of opportunistic bacterial infections among people living with HIV. Results About 18.06% (1824/9651) with (95% CI: 14.09–22.02) of the pooled population had opportunistic tuberculosis from 20 studies included, while from 16 included studies, the pneumonia infection was 11.64% (1040/8095/) with (95% CI: 8.45–14.83). Conclusion The prevalence of tuberculosis and pneumonia among people living with HIV in Ethiopia is high. Therefore, policymakers and health planners should put a great deal of emphasis on the implementation of relevant prevention and control measures. Registration The review was registered in the International Prospective Register of Systematic Reviews (PROSPERO) with the registration number “CRD42024587645”, on September 17, 2024.

Efficient elastic tissue motions indicate general motor skill