Browse Articles

Noise-robust markerless video gait anomaly detection via two-stage acquisition and LSTM autoencoders

Abstract We propose a markerless gait anomaly detection system for orthopedic screening. The framework combines MediaPipe-based joint tracking, unsupervised LSTM-autoencoder modeling, and targeted preprocessing to address clinical video noise. Trained only on normal gait, the model detects abnormal patterns in sarcopenia (SA) and Parkinson’s disease (PD) patients, achieving detection rates of 97% and 88%, respectively. Our method achieves state-of-the-art performance in sarcopenia detection, surpassing recent sensor-based approaches that require wearable devices or handcrafted features. Furthermore, to the best of our knowledge, this is the first unified markerless framework capable of identifying both sarcopenia and Parkinson’s disease using a single video-based system. To improve input quality, we address three common sources of error–frame imbalance, clothing interference, and background clutter—using YOLO-based frame filtering and semantic segmentation. This increased usable gait data by up to 38%. Joint-level analysis identified the knees as the most responsive to gait abnormalities, enabling interpretable and localized assessments. Our results highlight the potential of a scalable, non-invasive system for early detection and monitoring of musculoskeletal disorders.

Neural and endothelial cell-derived extracellular vesicles mediate Zika virus genome dissemination and productive infection in vivo

Zika virus (ZIKV) is a neurotropic flavivirus linked to severe neurodevelopmental defects following prenatal exposure. While the mechanisms by which ZIKV spreads within the central nervous system remain incompletely understood, extracellular vesicles (EVs) have emerged as potential mediators of intercellular communication and viral dissemination. Here, we demonstrate that EVs derived from ZIKV-infected neural cells encapsulate full-length viral genomes capable of establishing productive in vivo infection, independent of free virions. Primary cortical neurons, astrocytes, and mouse brain microvascular endothelial cells (MBECs) from neonatal mice were infected with ZIKV at a low multiplicity of infection (MOI 0.1). EVs were isolated and treated with acid glycine buffer and RNAase to exclude residual virions or free RNA. RNA sequencing, RT-qPCR, and droplet digital PCR (dd-PCR) analyses revealed that EVs—particularly those derived from neurons and MBECs—encapsulated ZIKV RNA, including full-length viral genomes. These EVs were able to transfer viral RNA to A549 cells in vitro , and its intracranial injection into neonatal mice resulted in productive infection, confirmed by detection of ZIKV capsid protein, viral RNA, and viral antigen in brain tissue. Our findings demonstrate that EVs from ZIKV-infected neural cells can serve as vehicles for genome transfer and initiate infection, even in the absence of detectable virions. The persistence of EVs-packaged genomes post-viremia could explain clinical observations of prolonged ZIKV RNA within the nervous tissue or delayed transmission. Understanding this pathway provides new insights into ZIKV neuropathogenesis and opens potential avenues for therapeutic intervention, for example targeting EVs biogenesis or cargo sorting.

Combined image block and DBSCAN processing for large field of view structured light point cloud denoising

Emerging insect-based aquafeed for sustainable African catfish production

The current study assessed the implications of substituting fish meal (FM) with black soldier fly larvae meal (BSFLM) on sustainable African catfish ( Clarius gariepinus ) production. Five isocaloric experimental diets were formulated and 168-days feeding trials were performed with 1400 improved strains of African catfish. Fish fed with 50% BSFLM expressed the lowest and highest feed conversion ratio, and daily weight gain and specific growth rate. Fillets from fish fed diet with 50 and 75% BSFLM inclusion expressed higher lauric (11.4–16.7%) and linoleic acids. Omega-6 fatty acid was 19-fold higher in fillets from fish fed diet with 75%BSFLM. There was 5–8% and 9–11% higher crude protein in fillet from fish fed diet with 50 and 75% BSFLM, respectively. Essential amino acids in the fillets, particularly lysine [11 – 18g/kg] and methionine [11 – 27g/kg] values were significantly higher in catfish fed diet with BSFLM diets than with FM. Diet integrated with 50% BSFLM showed the best return on investment (105.2%) and cost-benefit ratio (1.05). Integration of 50–75% BSFLM into African catfish diet offers a sustainable and cost-effective alternative to conventional fish feed, with promising long-term nutritional and health benefits for the aquaculture industry.

Remimazolam-induced anesthesia does not affect circadian rhythms significantly in mice

A nonsense mutation in the PRKG2 gene in dalmatian dogs with chondrodysplasia

Skeletal dysplasias encompass a diverse group of genetic disorders characterized by short stature and dwarfism. In humans, 771 types of skeletal dysplasia have been documented. Similar forms of these disorders have also been observed in dogs. The first cases of documented skeletal dysplasia in Dalmatian dogs were reported in the early 1980s, with additional affected dogs observed in subsequent years. Careful radiological and histopathological examinations at the time revealed severe limb deformities, including shortened radii and ulnae, irregular growth plates and disrupted endochondral ossification. In this study, we applied whole-genome sequencing on samples collected in 1992 and identified a genetic variant in the PRKG2 gene, introducing a premature stop codon (XM_038582312: c.1601T > G, p.L534X). Genetic variants in PRKG2 have previously been implicated in human acromesomelic dysplasia, a disorder affecting limb growth in young children. The PRKG2 -encoded protein plays a crucial role in endochondral ossification, and if translated, the identified nonsense variant would result in a truncated protein lacking most of the catalytic domain. Extended screening of the genetic variant revealed its continued segregation in the current Dalmatian population. Furthermore, three recent cases of dwarfism in Dalmatians were found to be homozygous for the identified PRKG2 nonsense variant. These findings provide compelling evidence for the role of PRKG2 in Dalmatian dwarfism, resolving a decades-old genetic mystery in the breed.

Reconstructing music perception from brain activity using a prior guided diffusion model

Glucose priming effect on microbial intercellular metabolic flux diversity in a marine intertidal sediment

Microbial communities inhabiting intertidal sediment show diverse metabolisms to adapt to hydrodynamic conditions. However, the priming effect of exogenous organic matter on microbial metabolic fluxes remains poorly understood. Here we investigated microbial intercellular activities in surface sandy intertidal sediment (0–5 cm depth, from the Nanhui tidal flat, East China Sea) under anoxic conditions by incubating with position-specific 13 C labelled glucose isotopologues. Elevated production of 13 C-CO 2 and phospholipid-derived fatty acids (PLFAs) was observed after 96-hour incubation, suggesting a positive priming effect on both catabolic and anabolic activities. The different incorporations of 13 C-label from glucose isotopologues into CO 2 and PLFAs reveal distinct intracellular carbon fluxes through the central metabolic network. Specifically, microbial communities preferentially utilize the Embden-Meyerhof-Parnas (EMP) pathway (62% flux), followed by the Pentose Phosphate (PP, 25%) and Entner-Doudoroff (ED, 13%) pathways. These flux distributions closely mirror their corresponding functional genome proportion (55% EMP, 26% PP, 20% ED), suggesting a metabolic balance between energy yield and enzyme cost across glycolytic routes. Furthermore, metabolic flux analysis based on 13 C-PLFA profiles highlights diverse metabolic strategies among different microbial taxa, primarily involving the EMP, ED and tricarboxylic acid pathways. This underscores the dual importance of energy production and carbon allocation for biomass synthesis. Given the high protein cost for the EMP pathway, energy acquisition may be prioritized by anaerobes to withstand the fluctuating conditions. Our results suggest that microbial interactions with intercellular networks may play a critical role in facilitating the priming and subsequent degradation of sedimentary organic matter.

Associations of adverse childhood experiences and depressive symptoms among children and adolescents: the mediating effect of resilience and sleep quality

Understanding real and mythical cancer risk factors: Insights from a university-based study

Despite extensive research on cancer etiology, many cancers remain associated with preventable risk factors. Raising public awareness of these factors is critical for promoting preventive behaviors. This study aimed to assess university students’ awareness of both scientifically validated and mythical cancer causes, to identify sociodemographic and health/lifestyle-related factors associated with awareness levels, and to examine the relationship between awareness of evidence-based and mythical cancer risk factors. A cross-sectional survey was conducted among 896 students from three Universities in Turkey. Data were collected using a Personal Information Form, the Cancer Awareness Measure (CAM) and Cancer Awareness Measure – Mythical Causes Scale (CAM-MYCS), which evaluates belief in widely held but scientifically unsubstantiated cancer causes (e.g., microwave ovens, stress, artificial sweeteners, electromagnetic frequencies). Statistical analyses included descriptive statistics, Spearman’s correlation, and multiple linear regression. Participants demonstrated significantly higher awareness of scientifically established cancer causes (70.7%) compared to mythical ones (10.2%). A moderate negative correlation was observed between CAM and CAM-MYCS scores (r = −0.511; p < 0.001), indicating that increased awareness of real causes is associated with decreased belief in myths. Regression analysis revealed that CAM scores were significantly predicted by academic year, being in the normal Body Mass Index (BMI) spectrum, and by having attended an oncology course (R² = 0.142; p < 0.001), whereas CAM-MYCS scores were predicted by academic year and high BMI values (R² = 0.072; p < 0.001). These findings highlight the need for targeted educational interventions aimed at both enhancing awareness of factual cancer risk factors and correcting persistent misconceptions. The results provide valuable insights for university-based public health initiatives seeking to improve cancer literacy and promote preventive health and lifestyle behaviors among young adults.

Implementation fidelity of universal design for learning and effects on student achievement engagement and belonging

Enhanced kernel search algorithm for optimizing local search capability and its application to carbon fiber draft process

Kernel Search Optimization (KSO) is characterized by insufficient accuracy in local search, which makes it difficult to achieve local optimization. Therefore, this paper proposes a Large Local Search Kernel Search Optimization (LLSKSO) to enhance the local optimization ability. LLSKSO achieves the performance improvement by introducing several strategies. First, the initial population is homogenized using the good point set mechanism. Then, the little dung beetle search mechanism of the Dung Beetle Optimizer (DBO) is introduced to enhance the local search capability of the KSO. Finally, the Cauchy-Gaussian mutation strategy is utilized to prevent the algorithm from falling into local traps. These three steps enable LLSKSO to achieve a dynamic balance between local and global search. In addition, to verify the performance and robustness of LLSKSO, comparison experiments between LLSKSO and 10 well-known algorithms are conducted on 50 benchmark test functions. From the statistical results of mean, best and variance of different algorithms, the LLSKSO algorithm outperforms the other algorithms. Finally, LLSKSO is applied to the engineering problem of carbon fiber drafting ratio optimization. Moreover, the experimental results obtained by LLSKSO yielded smaller line densities and greater strengths compared to other algorithms. LLSKSO achieves theoretical optima in 16 out of 20 high-dimensional benchmark functions, with an average CPU runtime reduced by 30% compared to baseline methods. Therefore, it can be shown that LLSKSO can be used as an effective optimization algorithm and engineering assistance.

Analysis and measurement of multi-factor coupling risk in underground coal mine gas explosion based on system dynamics

Polymer-coated urea effects on yield and gaseous loss of nitrogen for planting Chinese cabbage

The overuse of nitrogen fertilizer in Chinese vegetable fields, while boosting yield, is a major source of environmental pollution, particularly through N 2 O and NH 3 emissions. Optimizing nitrogen management is therefore crucial for reconciling productivity with environmental sustainability. Here, we assessed whether polymer-coated urea (PCU) with reduced application rates could sustain yield while minimizing environmental costs in Chinese cabbage production. Compared to conventional urea (260 kg N ha ⁻ ¹), PCU applications reduced by 0–20% increased nitrogen use efficiency by 47.7–49.9% and did not affect yield, whereas a 30% reduction caused significant yield loss. These PCU treatments also significantly reduced cumulative N 2 O and NH 3 emissions by 43.5–52.7% and 20.7–40.9%, respectively. We therefore recommend a 20% reduction in PCU-N application as the optimal management strategy for sustainable Chinese cabbage production, ensuring high yield with minimal environmental impact.

Adaptive course recommendation using federated learning and graph convolutional networks in IoT-enhanced e-learning

Occupational exposure to styrene and acute health effects among fiberglass-reinforced plastic workers: An integrated environmental and biological monitoring study

Styrene remains a major hazard in fiberglass-reinforced plastic (FRP) manufacturing. The current 10 ppm 8-h TLV-TWA is half the former limit, and the movement by several European states toward comparable or lower OELs highlights the need for fresh exposure–response data. In this study, eighty-five Korean FRP workers were monitored cross-sectionally. Full-shift breathing-zone styrene was measured by GC-FID; post-shift urine was analysed for mandelic acid (MA) and phenyl-glyoxylic acid (PGA), and dermal uptake was estimated with fluorescent tracers. Neuro-irritative symptoms were assessed by questionnaire and clinically verified in a subset. The results showed that median styrene levels were 18.65 ppm (spray-up), 12.42 ppm (hand lay-up) and 6.37 ppm (closed-mold). Urinary MA and PGA correlated with air levels (r = 0.78, 0.77). Dermal styrene load showed a moderate correlation with urinary MA (r = 0.42, p < 0.001). Symptom prevalence rose from 19% to 71% across exposure quartiles (adjusted OR = 5.6). A biomarker-based model using urinary mandelic acid (MA) with covariates (age, ventilation) showed strong apparent discrimination (AUC = 0.93). We propose 0.38 mg/g creatinine (MA) as a candidate operational (“early-warning”) threshold, pending external validation. In conclusion, integrated air, biological and dermal metrics reveal dose-dependent acute effects at or below 10 ppm. In this cross-sectional analysis, higher styrene exposure was associated with increased acute symptoms at or below ~10 ppm; these associations warrant confirmation in longitudinal studies with repeated biomonitoring. We present 0.38 mg/g creatinine (MA) as a candidate operational (“early-warning”)threshold to flag workers for closer evaluation; external validation is needed, and engineering controls remain the primary means of risk reduction.

Remnant cholesterol and body mass index jointly associate with cardiometabolic Multimorbidity with mediation analysis in CHARLS a prospective cohort study

What is a “Good” figure: Scoring of biomedical data visualization

Biomedical data visualization is critical for interpreting complex datasets, yet the clarity and quality of visualizations vary widely across tools and applications. This study introduces a comprehensive framework for evaluating biomedical figures and benchmarking visualization platforms. We developed Metrics for Evaluation and Discretization of Biomedical Visuals using an Iterative Scoring algorithm (M.E.D.V.I.S.), a quantification system that systematically assesses figure quality based on four criteria: Complexity, color usage, whitespace, and the number of distinct visualizations. The algorithm integrates dimensionality reduction, clustering, and thresholding to classify figures and generate tailored feedback for improvement. In parallel, we conducted a comparative analysis of 26 widely used visualization tools, evaluating each based on ease of use, customizability, financial cost, and required background knowledge. To demonstrate real-world applicability, we present case studies on figure evaluation in published research and introduce SpatioView, an interactive, browser-based platform for exploring spatial omics data. Collectively, our findings highlight the need for standardized evaluation methods and provide accessible solutions for improving figure design in biomedical research, education, and industry.

A linear programming joint optimization model of overnight train timetabling and maintenance planning on high-speed railway

Exploring the acceptability and attitude toward mobile health applications to aid self-management for people with type 2 diabetes mellitus in Saudi Arabia: A descriptive cross-sectional study

Background The prevalence of Type II diabetes mellitus (T2DM) is increasing, leading to a need for more sustainable and patient-centered solutions. The marketplace for mobile health (mHealth) applications (apps) that aid self-management of chronic diseases, such as T2DM, is growing. However, attitudes toward and acceptance of such technology in Saudi Arabia remain limited, particularly regarding preferences and needs. Therefore, understanding the acceptance and attitude toward using mHealth apps and preferred features is essential for planning appropriate methods and strategies to encourage usage among people with T2DM. This study aimed to determine the acceptance of and attitudes toward mHealth apps for self-management among people with T2DM in Saudi Arabia. It also seeks to understand the features that people with T2DM need and prefer regarding the mHealth apps. Methods A cross-sectional study using a structured self-administered questionnaire was conducted on people with T2DM living in Saudi Arabia. Data were analyzed using descriptive and inferential statistics to examine acceptance, perspectives, and attitudes toward using mHealth apps. Results A total of 424 individuals with T2DM completed the study questionnaire; among them, 92.5% owned smartphones, and 12.3% had previously used mHealth apps. Almost all respondents (98.1%) expressed willingness to use apps for self-managing T2DM. Medication reminders (78.3%) and diet management (69.8%) were the highest-ranked features of the mHealth app for T2DM self-management. Additionally, willingness to use apps for T2DM self-management was positively associated with education level and the number of medications taken daily (P = 0.002 and P = 0.04, respectively). Conclusion Overall, using a mHealth app appears to be an acceptable approach to aid the self-management of T2DM in Saudi Arabia, especially among the younger population. The features most required for future mHealth apps were identified, which could aid the design, acceptability, and usage of such apps.