Browse Articles
Discover research articles across all indexed journals
Early detection of occupational stress: Enhancing workplace safety with machine learning and large language models
Occupational stress is a major concern for employers and organizations as it compromises decision-making and overall safety of workers. Studies indicate that work-stress contributes to severe mental strain, increased accident rates, and in extreme cases, even suicides. This study aims to enhance early detection of occupational stress through machine learning (ML) methods, providing stakeholders with better insights into the underlying causes of stress to improve occupational safety. Utilizing a newly published workplace survey dataset, we developed a novel feature selection pipeline identifying 39 key indicators of work-stress. An ensemble of three ML models achieved a state-of-the-art accuracy of 90.32%, surpassing existing studies. The framework’s generalizability was confirmed through a three-step validation technique: holdout-validation, 10-fold cross-validation, and external-validation with synthetic data generation, achieving an accuracy of 89% on unseen data. We also introduced a 1D-CNN to enable hierarchical and temporal learning from the data. Additionally, we created an algorithm to convert tabular data into texts with 100% information retention, facilitating domain analysis with large language models, revealing that occupational stress is more closely related to the biomedical domain than clinical or generalist domains. Ablation studies reinforced our feature selection pipeline, and revealed sociodemographic features as the most important. Explainable AI techniques identified excessive workload and ambiguity (27%), poor communication (17%), and a positive work environment (16%) as key stress factors. Unlike previous studies relying on clinical settings or biomarkers, our approach streamlines stress detection from simple survey questions, offering a real-time, deployable tool for periodic stress assessment in workplaces.
Ensemble-based eye disease detection system utilizing fundus and vascular structures
Bulk and single-cell RNA-sequencing analyses revealed potential key genes and the role of CCL19/CCL21-CCR7 axis in hidradenitis suppurativa
Hidradenitis suppurativa (HS) is a chronic inflammatory skin disorder, affecting the pilosebaceous unit in apocrine gland-rich areas, characterized by painful nodules, abscesses and draining tunnels. The underlying molecular and immunological mechanisms remain poorly understood. This study aimed to identify key gene expression patterns, hub genes, and analyze the potential role of the CCL19/CCL21-CCR7 axis in HS lesions and peripheral blood using bulk and single-cell RNA sequencing analyses. By employing an integrative approach that included three machine learning methods and subsequent validation on an independent dataset, we successfully identified AKR1B10, IGFL2, WNK2, SLAMF7, and CCR7 as potential hub genes and therapeutic targets for HS treatment. Furthermore, our study found that CCL19 and CCL21 may originate from various cells such as fibroblasts and dendritic cells, playing a crucial role in recruiting CCR7-associated immune cells, particularly Treg cells. The involvement of the CCL19/CCL21-CCR7 axis in HS pathogenesis suggests that other CCR7-expressing cells may also be recruited, contributing to disease progression. These findings significantly advance our understanding of HS pathogenesis offer promising avenues for future CCR7-targeted therapeutic interventions.
Computational approaches in drug chemistry leveraging python powered QSPR study of antimalaria compounds by using artificial neural networks
Development of a Pediatric Vascular Catheterization Complication Score (Ped-VCCScore) for predicting post-cardiac catheterization complications
Cardiac catheterization, which is essential for clinical diagnosis and treatment, carries certain risks in pediatric patients, including complications such as loss of pulse, internal bleeding, vessel rupture, and subcutaneous hematoma. We investigated vascular complications in pediatric cardiac patients undergoing catheterization and developed a scoring system to predict these risks. We investigated pediatric patients aged <15 years who underwent cardiac catheterization at a tertiary hospital between January 2017 and December 2019 and developed a statistical model identifying the key factors influencing the risk of complications based on complication frequency, patient demographics, and treatment types. The identified key factors were body weight, procedure type, and maximum sheath size in the arterial-side-to-body-weight ratio. Using the scores determined by the model, participants were categorized into low-, intermediate-, and high-risk groups. The effectiveness of the model was assessed based on accuracy, alignment with real outcomes, and the ability to distinguish between cases with and without complications. Of the 390 patients, 6.2% experienced complications after cardiac catheterization. Transient pulse loss was the predominant complication (72%), followed by subcutaneous hematoma (12%) and bleeding (16%). In the development dataset, the vascular complication rates were 1.8%, 6.8%, and 26.1% in the low-, intermediate-, and high-risk groups, respectively. The likelihood ratios for vascular complications in the low-, intermediate-, and high-risk groups were 0.27 (95% confidence interval [CI]: 0.07, 0.81; P = 0.014), 1.12 (95% CI: 0.42, 2.67; P = 0.828), and 5.38 (95% CI: 2.23, 12.33; P < 0.001), respectively. Our model based on body weight, procedure type, and sheath size-to-body weight ratio accurately predicted vascular complications in pediatric cardiac catheterization. As one of the first studies to identify these risk factors, this study highlights the model’s potential applicability to support risk stratification-based clinical decision-making. Further validation in diverse clinical settings is needed to confirm its generalizability and predictive performance.
An experimental study examined how seawater depth affects the perormance of solar stills with and without silicon carbide
Best practices and practical strategies for co-designing virtual reality with Indigenous peoples: A scoping review protocol
Virtual reality (VR) is gaining traction in healthcare, education, and cultural sectors, from simulations in medical education to immersive museum experiences. Recently, VR has emerged as a powerful tool for Indigenous cultural preservation, language revitalization, and storytelling, offering immersive ways to safeguard knowledge and strengthen community connections. However, despite VR’s potential to support Indigenous self-determination, little is known about the extent of Indigenous leadership, engagements, and settler-Indigenous collaborations in VR development. There is a critical need to examine how VR can be ethically and meaningfully co-designed with Indigenous communities to ensure cultural integrity, respect for Indigenous knowledge systems, and equitable participation in technological innovation. Thus, this scoping review aims to identify practical strategies and best practices for co-designing VR with Indigenous communities. In accordance with the JBI methodology, we will conduct a comprehensive search across seven electronic databases, including MEDLINE (EBSCOhost), Scopus, Web of Science, ACM Digital Library, IEEE Xplore, Compendex (Engineering Village), and ProQuest Dissertations and Theses Global (ProQuest). Google Scholar will also be searched for grey literature sources. Eligible studies will focus on Indigenous populations (Population) and fully immersive VR co-design (Concept) across various contexts. Studies that do not discuss the design process will be excluded. Two independent reviewers will conduct literature screening, data extraction, and analysis, with findings synthesized narratively and presented in a structured charting table. The results will be disseminated through a peer-reviewed journal publication and shared with relevant community partners to support knowledge translation and application.
A simulation study on the temperature-dependent release of endocrine-disrupting chemicals from polypropylene and polystyrene containers
The critical role of timely medical emergency team activation in oncological and non-oncological patients
Purpose This study aimed to analyse the impact of time from first physiological derangement (positive Vital Sign Score, VSS) to Medical Emergency Team (MET) activation on mortality in oncological versus non-oncological patients. Methods All in-hospital patients requiring a MET assessment were included. The primary outcome was 30-day mortality. Subanalyses were performed for the population that was admitted to the ICU as well as the subgroups of oncological with solid vs. haematological cancer. In addition, we assessed the number and time points of VSS measurements in oncological and non-oncological patients. Results 286 out of 4068 (8.9%) documented MET calls were attributed to oncological patients. Each hour delay from first abnormal VSS to MET activation was associated with an 3% increase in adjusted 30-day all-cause mortality odds (OR 1.03, 95% CI 1.02–1.05, p < 0.001), independent of oncological status (interaction p = 0.141). Oncological patients had a significantly higher number of vital sign measurements on the ward (median 24 [16–37] vs 18 [10–27], p < 0.001) before MET activation, and higher 30-day all-cause mortality (38.1% vs 21.8%, p < 0.001). Conclusions Although the overall outcome did not differ between, oncological patients comprised 10% of MET calls, had higher mortality and had significantly more frequent vital sign monitoring, reflecting greater clinical concern. Tailored monitoring including early MET review has potential to improve outcomes particularly in high-risk oncological patients.
Author Correction: Effects of switching from intravitreal injection of aflibercept to faricimab on ocular blood flow in patients with diabetic macular edema
Interval type-2 intelligent fuzzy vehicle speed controller design using headlamp reflection detection and an adaptive neuro–fuzzy inference system
In this study, we present an algorithm to estimate the distance between a vehicle and a target object using light from headlights captured by a camera. In situations with limited distance data, we also design a fuzzy controller using the adaptive neuro–fuzzy inference system (ANFIS). To enhance robustness against disturbances, the interval type-2 approach is used. For the distance estimation algorithm, the vehicle is positioned at predefined intervals from the target object, capturing images of the headlights at each point. The region of interest containing the light is extracted from each image and segmented by light intensity. Weighted values are then assigned to each segment based on intensity, producing an image value that correlates with the distance. This image-derived value is then used as distance data for the design of the fuzzy controller. The controller is implemented using the interval type-2 fuzzy logic toolbox in MATLAB/SIMULINK, with vehicle speed and image intensity values as inputs and control torque as the output to adjust vehicle speed. The noise from the vehicle speed sensor is treated as a disturbance, and the performance of the interval type-2 fuzzy controller is evaluated under these disturbance conditions. Additionally, fuzzy controllers are designed for vehicle positions between 41–43 m and 47–49 m, and these controllers are trained using ANFIS to function effectively across the entire 41–49 m range. Simulation results demonstrate that, with the controller integrated into the vehicle system, the vehicle is successfully controlled to reach the target position.
New approaches to disinfection of thermolabile medical devices using an indirect method with cold atmospheric plasma-aerosol
Abstract Cold atmospheric plasma-aerosol (CAP-A) offers a promising alternative to conventional sterilisation and disinfection methods, which are often unsuitable for thermolabile medical devices due to high temperatures, toxic chemicals or radiation. CAP-A efficiently inactivates microorganisms and viruses without compromising the material integrity. Given the ongoing risk of infection associated with ultrasound probes and other delicate diagnostic instruments, this study investigates whether an indirect CAP-A method can meet all requirements for effective and safe disinfection of thermolabile medical devices. The disinfection of thermolabile medical devices was carried out in a container saturated with indirect CAP-A. A transvaginal ultrasound probe was used as a reference product. The study involved six test organisms, with five measurements taken at six different measurement points. The study showed that Enterococcus hirae (mean logarithmic reduction factor (LRF) > 6.23), Staphylococcus aureus (mean LRF > 6.51), and Enterococcus faecium (mean LRF > 6.16) demonstrated a germ reduction of > 99.9999%. For Pseudomonas aeruginosa (mean LRF > 5.40) and Escherichia coli (mean LRF > 5.29), a germ reduction of > 99.999% was achieved, and for Candida albicans (mean LRF > 4.95) and Clostridioides difficile (mean LRF > 4.62), a germ reduction of > 99.99% was demonstrated. The log reduction demonstrates a complete inactivation of the six tested microorganisms. The initially defined requirements for an effective disinfection process for thermolabile medical devices were met in the CAP-A method. Regarding highly tenacious microorganisms, such as Clostridioides difficile, the method of CAP-A proved effective, superior to alcohol-based methods, and with no resistance development observed. Its efficacy is otherwise only known in corrosive chemicals, such as hydrogen peroxide, chlorine, and chlorine dioxide. However, these chemicals have corrosive-oxidative effects on the surfaces to be disinfected and are critical in terms of market launch and hazardous material classification. Therefore, the method of CAP-A, provides an effective, material-friendly alternative.
Effectiveness of aerobic physical exercise on depression symptoms in adults: A protocol for developing a systematic review and meta-analysis of randomized clinical trials
Background Depression is a chronic condition that affects millions of people and requires effective interventions. Studies suggest that aerobic exercise can improve mental health and reduce depression symptoms, despite variations in exercise type and intensity. Objective This article presents a protocol for conducting a systematic review with meta-analysis, aimed at investigating the effects of aerobic exercise on reducing depression symptoms in adults. Methods This is a systematic review protocol, which will follow the PRISMA-P 2020 guidelines and has been registered in PROSPERO (no. CRD42024592700). The study search will be conducted in six databases: BVS, Cochrane library, Embase, Epistemonikos, PubMed and SPORTDiscus, using MeSH-based descriptors. Studies will be selected independently by two researchers using the Rayyan ‘QCRI’ software. Data extraction will be performed using specific forms, while the methodological quality of studies on physical exercise will be assessed with the TESTEX tool, and the risk of bias will be evaluated using the Cochrane RoB 2.0 method. The certainty of evidence will be evaluated through the Grading of Recommendations Assessment, Development, and Evaluation (GRADE). Conclusions It is expected that the systematic review following this protocol will identify the effective dose-response to reduce depression levels and provide an understanding of the mechanisms through which aerobic exercise influences depression.
Hybrid fuzzy logic approach for enhanced MPPT control in PV systems
Abstract This paper provides an in-depth analysis of photovoltaic (PV) system control within the MATLAB/Simulink environment, focusing on optimizing Maximum Power Point Tracking (MPPT) algorithms for enhanced efficiency under dynamic conditions. While conventional algorithms are widely used, their performance is limited under fluctuating conditions. To address this, we propose a novel hybrid approach combining Incremental Conductance with Fuzzy Logic Control (FLC), utilizing two innovative input variables: the sum of Conductance and Incremental Conductance (SInC) and its rate of change (CSI). The performance of the proposed algorithm, in comparison to other hybrid FLC methods, is evaluated through simulations using a boost converter under dynamic conditions, including abrupt irradiance changes and load variations. The results demonstrate that the proposed hybrid algorithm achieves superior performance, with an average MPPT efficiency of 97.7%, a convergence time of 53.5 ms, and an RMS of 97.8%, outperforming both conventional and other hybrid techniques. This work advances PV system control by providing a robust and adaptive solution for maximizing power extraction under diverse operating conditions.
Evaluation of sports training methods and technical characteristics based on multi-dimensional driving fuzzy intelligent computing
To improve the competitive state of badminton athletes and summarize the technical characteristics of badminton players, this paper introduces multi-dimensional fuzzy removal intelligent computing. Taking 120 badminton students from a sports school as data samples, the sports images of athletes are collected, the images are enhanced using histogram equalization, and then the fuzzy clustering algorithm is used to analyze the characteristics of the pictures. The following results were obtained from the analysis of the understanding degree of motion decomposition, the analysis of the lasting effect, the study of the number of repetitions, and the analysis of the simulation results: The degree of understanding was 17.75% higher than that of traditional training methods; the effect was better than that of conventional training methods; the traditional training method had a small number of action repetitions; the performance of boys and girls in the temporary mock exam would be related to different training methods. Therefore, this paper had practical significance for this research, to help promote such academic and give reference. At the same time, most optimization problems needed to comprehensively consider many factors, so multi-objective optimization algorithms became a hot spot in academic research.
Exploring the binding interaction of 1,4-naphthoquinone derivative–human serum albumin complex by biophysics and molecular simulation
Root pull-out resistance and surface microstructural characteristics of adapted plants in the water-level fluctuation zone of the three parallel rivers area
Root pull-out resistance is an important index to measure the soil-fixing ability of roots. The study aims to investigate the root pull-out resistance and root surface microstructural characteristics of plants adapted to the Water-Level Fluctuation Zone (WLFZ) and provide a reference for the study of vegetation soil reinforcement capacity in the WLFZ of the Three Parallel Rivers area. The study subjects are the four-year-old Arundo donax ‘Versicolor’, Cyperus involucratus, and Acorus calamus. The study employs the single root pull-out resistance experiments to determine their resistance. Additionally, SEM and paraffin sectioning methods were utilized to measure the microstructure of the root surface and to explore the differences in microstructure and their impact on the friction between the root and soil. The findings revealed (1) The failure modes of the single root pull-out experiments included both pull-out and breakage, with 70.83%, 81.48%, and 57.69% of the roots being broken for A. donax ‘Versicolor’, C. involucratus, and A. calamus, respectively. (2)There were significant differences in the average maximum pull-out resistance and average frictional strength among the three plants (P < 0.05), with the average maximum pull-out resistance being A. donax ‘Versicolor’ (27.88 N) > C. involucratus (20.53 N) > A. calamus (13.75 N), and the average frictional strength was A. donax ‘Versicolor’ (43.48 Pa) > C. involucratus (31.77 Pa) > A. calamus (19.05 Pa). The root surface roughness also showed significant differences among the three plants (P < 0.05), with the surface roughness of A. donax ‘Versicolor’ (20.13%) > C. involucratus (16.12%) > A. calamus (9.23%). (3) The root system of A. donax ‘Versicolor’ was relatively rough, with dense depressions and protrusions. In contrast, the root system of A. calamus was relatively smooth with no significant depressions or protrusions, and C. involucratus was intermediate between the two. The results suggested that the maximum pull-out force of single roots for the three plants followed the order of A. donax ‘Versicolor’ > C. involucratus> A. calamus. Moreover, the microstructure of the root surface had a significant impact on the maximum pull-out force of the roots, The rougher the root surface. The greater the single root drawing force.
Nanoparticles for improving biogas production and effluent biofertilizer
Abstract The impact of nanoparticles (NP) mixtures (Fe + Ni + Co, Fe + Ni, Fe + Co, and Ni + Co) on the anaerobic digestion (AD) of cow manure was examined through kinetic modelling (Modified Gompertz, Logistic, and First-order kinetic) and experimental observations. The NPs mixture significantly enhanced biogas production rate and yield potential. The variation between predicted and actual biogas yields was smaller with the Modified Gompertz model (1.37% to 5.30%) and First-order kinetic model (0.04% to 3.83%) compared to the Logistic model (1.99% to 7.04%). Additionally, digestates containing NPs showed strong fertility properties (5.16% to 5.36%). The combined use of kinetic models and experiments offers an effective way to quantify the influence of NPs mixture on AD performance, aiding in industrial application.
Exploring structural heterogeneity and organizational patterns of tourist flow networks among motivational subgroups
Tourists’ motivation is a key factor influencing their spatiotemporal behavior patterns. However, studies analyzing the spatial movement patterns of tourists with different motives are scant. In this study, the travel diaries of 4,306 tourists were employed as the primary data source, and the Louvain algorithm was applied to identify tourist flow networks with different motives; moreover, the complex network method was utilized to analyze the hierarchical structure and heterogeneity of the organizational patterns of tourist flow networks with different motives. The analysis revealed the following main conclusions. (1) There are three types of tourism flow networks with different motivations: exploring nature, exploring culture and nature-culture motivations. (2) All types of motivated tourist flow networks exhibit significant node hierarchies. The core nodes in natural network are the most controllable, whereas those in natural-cultural network are less controllable and more evenly connected. (3) Node connectivity strength decays significantly with distance. Geographic constraints exert the greatest impact on natural network. (4) Different-motivated tourist flow networks present a Multi-scale Hierarchical Nested Pattern. Tourists gather in multi-core cultural networks and multi-core natural-cultural networks, using them as a base to spread to neighboring regions. They shift between core nodes and use new cores as a base for tours. This study enriches the theoretical system of tourist behavior research and provides effective practical recommendations for the Shanxi tourism industry to improve tourists’ travel experience.