Browse Articles

A new dataset for measuring the performance of blood vessel segmentation methods under distribution shifts

Creating a dataset for training supervised machine learning algorithms can be a demanding task. This is especially true for blood vessel segmentation since one or more specialists are usually required for image annotation, and creating ground truth labels for just a single image can take up to several hours. In addition, it is paramount that the annotated samples represent well the different conditions that might affect the imaged tissues as well as possible changes in the image acquisition process. This can only be achieved by considering samples that are typical in the dataset as well as atypical, or even outlier, samples. We introduce VessMAP, an annotated and highly heterogeneous blood vessel segmentation dataset acquired by carefully sampling relevant images from a large non-annotated dataset containing fluorescence microscopy images. Each image of the dataset contains metadata information regarding the contrast, amount of noise, density, and intensity variability of the vessels. Prototypical and atypical samples were carefully selected from the base dataset using the available metadata information, thus defining an assorted set of images that can be used for measuring the performance of segmentation algorithms on samples that are highly distinct from each other. We show that datasets traditionally used for developing new blood vessel segmentation algorithms tend to have low heterogeneity. Thus, neural networks trained on as few as four samples can generalize well to all other samples. In contrast, the training samples used for the VessMAP dataset can be critical to the generalization capability of a neural network. For instance, training on samples with good contrast leads to models with poor inference quality. Interestingly, while some training sets lead to Dice scores as low as 0.59, a careful selection of the training samples results in a Dice score of 0.85. Thus, the VessMAP dataset can be used for the development of new active learning methods for selecting relevant samples for manual annotation as well as for analyzing the robustness of segmentation models to distribution shifts of the data.

Targeted analysis of dyslexia-associated regions on chromosomes 6, 12 and 15 in large multigenerational cohorts

Dyslexia is a common learning impairment with a genetic basis that affects word reading and spelling. An increasing list of loci and genes have been implicated, but analyses to-date have investigated only limited genomic variation within each locus with no confirmed pathogenic variants identified. Our study is the first to comprehensively sequence both coding and cis-acting regulatory regions of such genes in a large study sample. In a collection of >2000 participants in families from three independent sites, we performed targeted capture and comprehensive sequencing of all exons and some regulatory elements of five candidate risk genes ( DNAAF4 , CYP19A1 , DCDC2 , KIAA0319 and GRIN2B ) for which prior evidence for a role in dyslexia exists from more than one sample. We evaluated evidence for association in each of six dyslexia-related quantitative phenotypes (traits) using both individual common single nucleotide polymorphisms and aggregated rare variants. We detected no promoter alterations and few deleterious variants in the coding exons, none of which showed evidence of association with any trait. Single variant and aggregate testing of DNAAF4 failed to detect significant evidence of association with any of the traits. The other four genes provided evidence of association with one or more traits. A common variant downstream of CYP19A1 showed significant evidence of association with multiple traits with or without verbal IQ (VIQ) adjustment. A haplotype that stretches from the downstream region of KIAA0319 to the second intron of DCDC2 was associated with reduced performance on timed real word reading. Finally, rare exonic variants in GRIN2B were associated with performance on spelling, with or without adjustment for VIQ. Our findings from this large-scale sequencing study complement those from genome-wide association studies, argue against the causative involvement of large-effect coding variants in these five candidate genes, support a multigenic etiology, and suggest a role of transcriptional regulation.

Drivers and forecasting of carbon emissions with extended LMDI and Bagging models: A case study of China’s Bohai Rim region

The Bohai Rim region, one of the prominent coastal economic zones in China, has garnered considerable attention regarding its carbon emission characteristics and driving factors. This paper computes the overall carbon emissions within the Bohai Rim by utilizing economic and energy usage data spanning the years 2009to 2021. The results indicate a clear pattern in carbon emissions, which rise at first before experiencing a decline, primarily driven by substantial contributions from the industrial and transportation sectors. The three-layer extended Log Mean Divisia Index (LMDI) decomposition model examines how different driving characteristics affect carbon emissions, including energy composition, industrial framework, and the level of economic advancement. The study finds that despite regional economic growth, carbon emissions are declining overall. In the industrial sector, emissions reduce in large part by optimizing the energy mix and increasing energy efficiency. Similarly, modal shifts and energy efficiency advancements positively impact emission reduction in the transportation sector. Furthermore, trends in carbon emissions are predicted using a bagging ensemble model for the Bohai Rim region’s two municipalities and three provinces from 2024 to 2060. This study also examines the carbon emission reduction potential under high-speed, low-speed, and normal socio-economic development scenarios. These findings offer an in-depth insight into carbon emissions in the Bohai Rim of China and provide theoretical foundations and policy suggestions aimed at fostering low-carbon growth in related coastal zones.

Reconstructing smoking history through dental cementum analysis - a preliminary investigation on modern and archaeological teeth

Acellular extrinsic fibre cementum (AEFC) has been widely utilised in cementochronology to estimate age at death, seasonality, and for life-history reconstruction. Smoking has been commonplace in the UK since the 17th century and is known to compromise oral health and to modulate physiological processes. This study aimed to investigate whether AEFC analysis could identify smoking activity in both modern and archaeological populations. A modern sample (70 teeth from 46 donors) with known age, sex, and smoking status was compared with an archaeological sample (18 teeth from 18 individuals), dating from the 18th/19th centuries in Coventry, UK, whose biographical information was recorded from coffin plates where available. Smoking status for the archaeological individuals was inferred from pipe notches and dental staining. AEFC analysis that was blinded to smoking status measured increment count, overall width and the presence of irregularities within the cementum microstructure in both samples. Results demonstrated that the AEFC width was significantly lower (p = 0.008) in current smokers compared to ex-smokers. Additionally, individuals with a history of smoking were significantly more likely to display disrupted incremental patterns within their AEFC (p < 0.001). This research suggests an association between smoking and periodontal ligament health, which influences AEFC formation and shows that the AEFC provides a record of smoking-related oral health damage. This research expands the potential applications of cementochronology to forensic and archaeological investigations for life history reconstruction.

Assessing social cognition and risk-taking behaviour in patients with young-onset dementia: Study protocol for the YOD-RiSoCo observational prospective cohort study

Background Certain subtypes of young onset dementia (YOD), such as the behavioural variant of FTD or the behavioural variant of AD (bvYOD), present with changes in social behaviour instead of memory impairments. These symptoms are often under-recognized, delaying the diagnosis and contributing to psychosocial problems. Impairments in social cognition (SC), an important affected domain in bvYOD, underlie these social behavioural changes. Especially emotional blunting and a lack of empathy in patients with bvYOD might be related to problematic social behaviour, such as risk-taking behaviour, which may potentially harm others. However, despite the importance of SC impairments in the diagnosis of YOD and the impact of SC impairments on social behaviour, there is a lack of valid and well normed measures for certain aspects of SC, such as emotion experience and empathy. Methods The YOD-RiSoCo study is an observational prospective cohort study, consisting of two separate, but related, studies. Study 1 includes 64 patients with bvYOD and 64 healthy controls to assess the sensitivity and validity of newly developed SC instruments for measuring emotion experience and empathy, by comparing their average group performance. Furthermore, validity of the new instruments will be assessed by analysing the associations of performances on these new tests with those on more traditional SC and other neurocognitive tests. Study 2 focuses on assessing to which extent SC measures relate to risk-taking behaviour. This study includes 20 patients with bvYOD and 20 healthy controls from Study 1, in addition to 20 patients with non-bvYOD (e.g. Alzheimer’s dementia or vascular dementia) and 20 patients with serious brain injury affecting frontal networks. A specific question is whether the relationship between SC and risk-taking behaviour is generic (for all groups with SC impairments), or specific (not in dementia without SC impairments). Discussion Results of the YOD-RiSoCo study will yield new, sensitive neuropsychological tests for aspect of social cognition, which may contribute to a more timely diagnosis of YOD, allowing earlier provision of appropriate counselling and care for patients and their close others. Furthermore, the study will contribute to a better identification of those social behavioural symptoms that negatively affect functioning and social relations. Trial registration The trial is registered at www.clinicaltrials.gov with identifier NCT06286293

GNN-FTuckER: A novel link prediction model for identifying suitable populations for tea varieties

Current research on tea primarily focuses on foundational studies of phenotypic characteristics, with insufficient exploration of the relationship between tea varieties and suitable populations. To address this issue, this paper proposes a link prediction model based on Graph Neural Networks (GNN) and tensor decomposition, named GNN-FTuckER, designed to predict the “tea suitability” relationships within the tea knowledge graph. This model integrates the SE-GNN structural encoder with an improved TuckER model decoder. The SE-GNN encoder enhances the modeling capability of the global graph structure by explicitly modeling relations, entities, and triples, thereby obtaining embedding vectors through aggregation, updating, and iterative operations. The improved TuckER model enhances the capture of complex semantics between entities and relations by introducing nonlinear activation functions. To support our research, we constructed a tea dataset, TeaPle. In comparative experiments, GNN-FTuckER achieved superior performance on both public datasets (WN18RR, FB15k-237) and the TeaPle dataset. Ablation studies indicate that the model improved H@10 by 4.3% on the WN18RR dataset and by 1.5% on the FB15k-237 dataset, with a 1.3% increase in MRR. In the TeaPle dataset, H@3 improved by 4.7% and H@10 increased by 3.1%. This research provides significant insights for further exploring the potential of tea varieties and evaluating the health benefits of tea consumption.

Screening of agronomic and qualitative physical and chemical traits of 83 naked oats strains

Naked oats (Avena nuda) are a key cereal crop that are valued for their high nutritional content and desirable agronomic traits. In this study, 80 developed strains and three bred varieties were evaluated in Kelan County, Shanxi Province, China, from 2022 to 2023. Six agronomic, eight physical, and five chemical traits were assessed. Significant phenotypic variations were observed across the 83 strains, with marked improvements in agronomic traits in 2023, particularly in grain shape, thousand grain weight, and spike length. The grain shape showed the most substantial improvement, whereas the thousand grain weight remained relatively stable. Cluster analysis identified three agronomic groups, with Group III comprising high-yield, stable strains, such as YM49. Regarding the physical traits, a general reduction in grain size and fullness was observed in 2023, with hardness exhibiting the greatest variation. Principal component analysis classified the strains into three morphological groups, with YM23 representing a large-grain, soft-textured type. Chemical quality traits indicated an increase in β-glucan and water content in 2023, whereas fat content decreased while exhibiting high variability. Nutritional profiling clustered the strains into three groups, with YM39 leading in protein, β-glucan, and fat content. Correlation analysis across the selected 19 traits highlighted the key tradeoffs between grain morphology, yield, and nutritional quality. Effective tillering was identified as a crucial trait that influences the overall performance. Strain YM49 demonstrated a consistent high-ranking performance across multiple traits, indicating its potential for breeding high-quality oats with high yields.

Repetitive Transcranial Magnetic Stimulation targeted with MRI based neuro-navigation in major depressive episode: a double-blind, multicenter randomized controlled trial

Context High-frequency (HF) transcranial magnetic stimulation (rTMS) of the left dorsolateral prefrontal cortex (DLPFC) is widely used in Major Depressive Episode (MDE). Optimization of its efficacy with a neuro-navigation system has been proposed based on a small randomized controlled trial (RCT) supporting a large effect. Method This evaluator- and patient-blind, multicenter RCT assessed the superiority in terms of efficacy of 10 HF rTMS sessions of the left DLPFC targeted with MRI based neuro-navigation versus similar sessions targeted by the standard 5 cm technique. The study was conducted between January 2013 and April 2017, at 4 hospitals centers in France where both in- and out- patients with MDE were included. Randomization was computer-generated (1:1), with allocation concealment implemented within the e-CRF. The main outcome measure was the percentage of responders 44 days (D44) after the rTMS session. Secondary outcomes were percentage of remitters, Beck Depression Inventory and psychomotor retardation assessed with Salpêtrière retardation rating scale (SRRS) for depression at D14 and D44. The results are presented along with their 95% confidence intervals. Results 105 patients were randomized and 92 were evaluable with respectively 45 patients in the neuronavigation group and 47 in the standard group. A treatment response was observed for 14 (31.8%) of 44 patients analyzed in the intervention group, and for 16 (35.6%) of 45 patients analyzed in the control group with no statistical difference (relative risk 0.89; 95% confidence interval, [0.50;1.61]). No difference was evidenced for secondary outcomes at D44 whether it concerns remission at D44 (relative risk, 0.82; 95% CI, 0.36 to 1.88), or BDI results (difference in means, 0,01; 95% CI, -3.06 to 3.26), or SRRS results (difference in means, 0.11; 95% CI, -2.42 to 5.02). Similar results were observed at D14. Rates of adverse events were similar in both groups with 23 (47.9%) and 1 (2.1%) of adverse events and serious adverse events in the neuro-navigation group versus 20 (40.8%) and 0 (0%) in the standard group. Discussion This study failed to reproduce previous findings supporting the use of neuro-navigation system to optimize rTMS efficacy. Limitations of this study includes a small sample size and a number of rTMS sessions that may appear substandard in 2025. Trial registration: NCT01677078

The efficacy of Netrod Six-electrode Radiofrequency Renal Denervation system: mathematical modeling, phantom experiments and animal studies

The principle of NetrodTM Renal Denervation RF Generator is based on the heating of radiofrequency electrodes to achieve the therapeutic temperature for peripheral nerve ablation with minimal impact on intra-arterial blood flow. The extravascular heat distribution and continuous ablation were evaluated using COMSOL Multiphysics with an idealized 3D model and validated in an ex vivo experiment. The safety of ablation was verified by H&E staining of a hypertensive dog model in the treatment group of animal experiments, and changes in blood pressure before and after the procedure(assessed at 7, 30, and 60 days in animal experiments) were used to characterize the efficacy. The simulation results showed that the maximum injury depth range of the NetrodTM Six-electrode is 5.08mm and the depth of the ablation zone obtained in the Phantom experiment was 3–6mm. Circumferential coverage was up to 91.99% and the ablation zone was continuous. In animal experiments, eight animals in the treatment group had significantly lower blood pressure (88.9%). Nerve cells in nerve bundles showed vacuolar changes and thickening of peripheral sheaths after peripheral nerve ablation of renal arteries.The NetrodTM Six-electrode presented here has potential to effectively and safely ablate renal nerves. The demonstrated efficacy and safety in animal models provide a foundation for clinical trials in humans.

Fractional order tracking control of a disturbed differential mobile robot

A trajectory tracking problem for a wheeled mobile robot, under skidding and slipping effects, is addressed. A sliding mode control scheme with fractional order reaching dynamics is proposed to solve the problem. The design of the control scheme is based on backstepping and passivity feedback equivalence techniques. The skidding and slipping effects are treated as a part of exogenous disturbances associated to a kinematic model of the robot. A formal study, based on Lyapunov stability theory, is carried out to assure the existence of a sliding surface where trajectory tracking is obtained; the same stability tool is used to study the trajectory tracking error behavior of the closed loop system. The performance of the control scheme is evaluated by means of numerical simulations and experimental results showing a good performance when compared with the corresponding integer order control scheme.

Methanolic crude extract of Litsea Monopetala leaves combats oxidative stress, clot formation, inflammation and stool frequency in animal model

Introduction Litsea monopetala (LM) leaves are used in traditional medicine system in the South Asian region for treating ailments such as digestive issues, respiratory problems and skin disorders. In this study, we investigated the possible antioxidant, thrombolytic, analgesic and antidiarrheal properties of the methanolic extract of LM leaves. Method We assessed the antioxidant activity using DPPH (2,2-diphenyl-1-picrylhydrazyl) free radical scavenging and total phenolic content tests, while thrombolytic activity was evaluated by clot lysis assays. The in-vivo analgesic and antidiarrheal activities were tested by two standard methods, e.g., acetic acid-induced and castor oil-induced animal model, respectively. Prior to in vivo and in vitro evaluation of the pharmacological activities phytochemical screening was also performed to estimate the bioactive compounds (e.g., phenol, carbohydrates, reducing sugars, tannins, alkaloids, flavonoids, saponins and steroids) present. Results Among the tested phytochemicals, our results reveal carbohydrates, alkaloids, flavonoids and tannins as the major phytocompounds present in the extract. Within the different solvent extractives, the methanolic extract exhibited DPPH free radical scavenging features with an IC50 of 8.99 µg/ml compared to ascorbic acid, an IC50 of 13.38 µg/ml. At 500 mg/kg dose, the extract produced 67.05% decrease in the frequency of acetic acid-induced writhing while diclofenac sodium showed decrease by 74.25%. The extract also significantly (P < 0.01) decreased the frequency of castor oil-induced diarrhea in compared to the standard drug of loperamide. Finally, the clot lysis assay with the methanolic extract demonstrated an increase in the thrombolytic activity by 40.79% compared to streptokinase, which increased by 69.52%. Conclusion Overall, this study shows promise that the methanolic crude extract of LM leaves may contribute to the alternative or additive strategy to modulate conditions of oxidative stress, thrombolytic, inflammation and diarrhea. Further comprehensive investigation is necessary to clarify the exact mechanisms of action and the phytochemical composition of LM leaves.

Effect of switching from prior Nucleos(t)ide Analogue(s) to Tenofovir alafenamide on lipid profile and cardiovascular risk in patients with Chronic Hepatitis B

Introduction Tenofovir alafenamide (TAF) is recommended for chronic hepatitis B (CHB) treatment in international guidelines according to its efficacy and safety. However, in phase III study, an increased LDL-c was observed in those who were switched from Tenofovir disoproxil fumarate (TDF) to TAF. Limited data exists on whether lipid profiles change only in individuals who switched to TAF from TDF or from any nucleoside/nucleotide analogues (NUC). We investigated how switching to TAF affected lipid and cardiovascular outcomes in Thai CHB patients. Materials and methods We conducted a prospective observational study including CHB patients who had to switch from their prior NUC to TAF according to the national reimbursement policy in late 2022. All enrolled patients had lipid tests and transient elastography (TE) done at 0 and 48-week post-switch to TAF. Demographic data, prior NUC, liver biochemistry, controlled attenuated parameter (CAP) and liver stiffness (elastic modulus; E) data measured by TE were collected. The changes in lipid, Thai cardiovascular (CV) risk score, and TE results between 0 and 48-week were compared. Results A total of 110 patients who were switched to TAF and completed 48-week follow-up were analyzed. The prior NUCs were as follows: 47 Lamivudine (LAM), 22 Entecavir (ETV), and 41 TDF-based. Baseline characteristics were similar between the three groups except for underlying hypertension was more frequent and baseline total cholesterol was lower in the TDF-based group. At 48-week post-switch, the median LDL-c changes were -2.45, -5.9 and +8.8 mg/dL (p<0.001), and total cholesterol changes were -4.5, -4 and +17 mg/dL (p<0.001), in the ETV, LAM, and TDF-based group, respectively. Whereas the changes in hepatic steatosis (measured by CAP), and liver stiffness (measured by E) as well as Thai CV risk score were not significantly different. No cardiovascular events occurred during follow-up. Conclusion Significant increase in LDL-c and total cholesterol after switching to TAF were observed only in patients with prior TDF, but not in those with prior ETV or LAM. Careful monitoring of lipids after the switch may not be universally needed. Data regarding long-term cardiovascular outcomes are warrant.

Proteomes of aging and omega-3 supplementation in rat soleus skeletal muscle

Aging is a pan-organ process with an intricate and multimodal nature. Deciphering the aging phenomenon is complex, yet recent attention is analyzing the potential benefits of non-invasive life adjustments to achieve healthy aging. Omega-3 polyunsaturated fatty acids (FA) have emerged as promising nutraceuticals for a plethora of different medical conditions. In the current study we conducted an in-depth, bottom-up, global, shotgun proteomic study (LC-MS/MS) investigating both the effects of aging on skeletal muscle and the potential alterations due to ω-3 FA. Sprague Dawley rats were fed different diets and divided into four groups (n = 5 per group): adult controls (7–8 months, ADCTL); aged controls (22 months, AGCTL); and adult (ADω3) and aged (AGω3) rats fed an ω-3 supplemented diet. Among the identified 30,000 soleus proteins, our proteomic analysis identified 149 proteins differentially expressed in aging; 207 proteins with aging, but fed ω-3 FA; and 105 and 26 proteins, respectively, when aged and adult rats were fed ω-3 FA. Aging alone (ADCTL/AGCTL) affects many processes: carbohydrate and lipid metabolisms, proteostasis, mRNA processing and sarcomeric proteins. With FA supplementation and aging (ADω3/AGω3) similar processes were affected, but increased chromatin-related protein abundances (methylation or histone deacetylation) were observed in AGω3; while proteins involved in OXPHOS and mitochondrial homeostasis, including mTOR, were more represented in ADω3 rats. Supplementation with FA had a greater effect in aged rats (AGCTL/AGω3) than in adult ones (ADCTL/ADω3). In the ADCTL/ADω3 comparison, modest changes were seen, whereas in the AGCTL/AGω3 comparisons DNA damage repair increased and protein synthesis and degradation were observed. Further, a potential link to enhancement of myogenesis is also evident. The data presented in this work suggest potential beneficial and protective effects of ω3 FA supplementation in the soleus muscle, as well as some potential molecular mechanisms of action.

Retraction: Clinical features of 375 COVID-19 cases imported from Russia through the Suifenhe port and countermeasures

Functional safeguards for conservation: Identifying climate change refugia for frugivorous and nectarivorous birds in a degraded area of Colombia

Habitat loss and climate change are major drivers of biodiversity loss, but their synergistic effects and functional perspectives have to be better understood. We employed species distribution models under future contrasting socioeconomic scenarios to assess the impacts of climate change and human footprint on avian frugivore and nectarivore functional groups in the Magdalena Valley, a highly transformed and biodiverse region in Colombia. We constructed the functional groups based on a dissimilarity matrix with 16 anatomical and ecological traits. Two types of future climatic refugia (type 1: areas that will maintain the current climatic conditions and type 2: regions outside the current distribution area that will have the current climatic conditions) were identified to guide conservation efforts for these groups and associated ecosystem services. Of the 27 functional groups identified, 19 are projected to undergo range reductions of 1–75%, with an average upward shift of their climatic niches along the altitudinal gradient of 690 m. Large frugivores from intermediate elevations, such as toucans and cracids, as well as nectarivores with extreme adaptations and specializations, are expected to experience the most severe range reductions. Distributional and altitudinal shifts will lead to spatial reorganization of communities and a reduction or complete loss of functional group richness, particularly in lowland areas. This could impact ecosystem services relevant for degraded area restoration, such as seed dispersal, fruit availability, and pollination of specialized plant species with economic importance. The low representation of future climatic refugia within protected areas highlights the need to incorporate climate change trends into future conservation strategies for these landscapes.

An Interval-valued intuitionistic Fuzzy group decision-making approach based on projection measurement extended VIKOR for patent quality evaluation

Patent quality assessment is a multi-attribute decision-making problem. The increasingly complex social environment makes group decision-making based on interval-valued intuitionistic fuzzy (IVIF) information more and more important. In group decision-making, considering using a comprehensive projection measure to calculate the similarity between two decision matrices is one of the important steps. In order to make up for the deficiencies of the classical projection measure and the existing normalized projection measure in measuring the similarity between two IVIF vectors and optimize the group decision-making performance of the VIKOR method, this study proposes: (1) a new definition of subtraction of IVIF numbers, which further enriches the theoretical basis of IVIF numbers; (2) a new normalized projection measure to measure the similarity between two IVIF matrices; (3) combined with the regret matrix to store group regret information, which enhances the comprehensive ability of the classical VIKOR technique, and then establishes a comprehensive GDM method for patent quality assessment. Finally, a specific patent quality assessment example in the existing literature is used for data experiment. Through static and dynamic experimental comparative analysis, the feasibility and effectiveness of the proposed method are illustrated, and it can almost produce a stable ranking. The research in this paper not only provides an effective group decision-making tool for patent quality assessment, but also provides theoretical support and practical guidance for fuzzy group decision-making problems in other fields.

Small means immoral? The impact of spatial size metaphor on moral judgment

This study aims to explore the unconscious relationship between moral concepts and the spatial dimension of size, as well as to examine whether the unknown size of a room influences participants’ moral cognitive judgments within the framework of embodied cognition. Study 1 and Study 2 investigate participants’ unconscious biases. Specifically, participants exhibited faster response times when judging moral concepts presented in large fonts and sizes and immoral concepts presented in small fonts and sizes, compared to when moral concepts were presented in small fonts and sizes and immoral concepts in large fonts and sizes. Study 3 employed a moral dilemma task, revealing that participants placed in a large room evaluated characters in a story more morally under the embodiment effect than those in a small room. Collectively, these three studies demonstrate that the unconscious psychological relationship between moral concepts and the spatial dimension of size influences individuals’ abstract moral judgments under embodied cognition.

InBRwSANet: Self-attention based parallel inverted residual bottleneck architecture for human action recognition in smart cities

Human Action Recognition (HAR) has grown significantly because of its many uses, including real-time surveillance and human-computer interaction. Various variations in routine human actions make the recognition process of action more difficult. In this paper, we proposed a novel deep learning architecture known as Inverted Bottleneck Residual with Self-Attention (InBRwSA). The proposed architecture is based on two different modules. In the first module, 6-parallel inverted bottleneck residual blocks are designed, and each block is connected with a skip connection. These blocks aim to learn complex human actions in many convolutional layers. After that, the second module is designed based on the self-attention mechanism. The learned weights of the first module are passed to self-attention, extract the most essential features, and can easily discriminate complex human actions. The proposed architecture is trained on the selected datasets, whereas the hyperparameters are chosen using the particle swarm optimization (PSO) algorithm. The trained model is employed in the testing phase for the feature extraction from the self-attention layer and passed to the shallow wide neural network classifier for the final classification. The HMDB51 and UCF 101 are frequently used as action recognition standard datasets. These datasets are chosen to allow for meaningful comparison with earlier research. UCF101 dataset has a wide range of activity classes, and HMDB51 has varied real-world behaviors. These features test the generalizability and flexibility of the presented model. Moreover, these datasets define the evaluation scope within a particular domain and guarantee relevance to real-world circumstances. The proposed technique is tested on both datasets, and accuracies of 78.80% and 91.80% were achieved, respectively. The ablation study demonstrated that a margin of error value of 70.1338 ± 3.053 (±4.35%) and 82.7813 ± 2.852 (±3.45%) for the confidence level 95%,1.960σx̄ is obtained for HMDB51 and UCF datasets respectively. The training time for the highest accuracy for HDMB51 and UCF101 is 134.09 and 252.10 seconds, respectively. The proposed architecture is compared with several pre-trained deep models and state-of-the-art (SOTA) existing techniques. Based on the results, the proposed architecture outperformed existing techniques.

Efficacy and safety of acupuncture for early postpartum stress urinary incontinence: A protocol for a pilot randomized controlled trial

Objective Postpartum stress urinary incontinence (PSUI) is a common condition among women after childbirth. While acupuncture is a common clinical treatment for PSUI, high-quality clinical evidence supporting its effectiveness is currently lacking. This study aims to preliminarily evaluate the efficacy and safety of acupuncture for early postpartum stress urinary incontinence, providing reference data for sample size calculation and protocol feasibility in formal trials. Design and methods This is a randomized controlled trial. Seventy-two PSUI patients between 42 days and 1 year postpartum will be randomly assigned in a 1:1 ratio to either the acupuncture group (n = 36) or the sham acupuncture group (n = 36). Both groups will receive acupuncture treatments three times per week for two weeks. Both groups will receive identical education regarding pelvic floor muscle training. The primary outcomes are changes in urine leakage volume measured by 1-hour pad test at week 2 of treatment and week 6 of follow-up compared to baseline (week -1), and the proportion of patients achieving at least 50% reduction in urine leakage volume at week 2 compared to baseline. Secondary outcomes include 72-hour bladder diary, The International Consultation on Incontinence Questionnaire-Short Form (ICIQ-SF), Incontinence Quality of Life questionnaire (I-QOL), and pelvic floor muscle strength assessment. Discussion The results of this study will provide preliminary evidence on the efficacy and safety of acupuncture for PSUI, offering reference data for sample size calculation and protocol feasibility in formal trials, ultimately providing more treatment options for PSUI patients. Trial registration ITMCTR2024000290.

Global, regional, and national burden of osteoarthritis from 1990 to 2021 and projections to 2035: A cross-sectional study for the Global Burden of Disease Study 2021

Objective This study aims to report the trends and cross-national disparities in the burden of osteoarthritis (OA) by region, age, gender, and time from 1990 to 2021, and to further project changes through 2035. Methods In this systematic analysis based on the Global Burden of Disease (GBD) study, population survey data on osteoarthritis from 21 countries/regions and U.S. insurance claims data were used to estimate the prevalence and incidence of OA in 204 countries and regions from 1990 to 2021. The reference case definition for OA was symptomatic and radiographically confirmed osteoarthritis. Studies using definitions other than the reference, such as self-reported OA, were adjusted through a regression model to align with the reference case. The distribution of OA severity was derived from a pooled meta-analysis using the Western Ontario and McMaster Universities Arthritis Index (WOMAC). Final prevalence estimates were multiplied by disability weights to calculate years lived with disability (YLD). An Autoregressive Integrated Moving Average (ARIMA) model was used to forecast the prevalence and incidence of OA through 2035. Results In 2021, approximately 607 million (95%UI 538–671) people worldwide were affected by osteoarthritis, accounting for 7.7% of the global population. Compared to 2020, the age-standardized prevalence of OA among males is projected to increase from 5,763 per 100,000–5,922 per 100,000 by 2036, while the age-standardized prevalence among females is expected to decline slightly from 8,034 per 100,000–7,925 per 100,000. In 2021, the global age-standardized YLD rate for osteoarthritis was 244.5 (95%UI 117.06–493.11), the global age-standardized prevalence rate was 6,967.29 (95%UI 6,180.7–7,686.06), and the global age-standardized incidence rate was 535 (95%UI 472.38–591.97). In 2021, the age-standardized prevalence rate exceeded 5.5% across all regions, ranging from 5,675.8 per 100,000 (95%UI 5,001.76–6,320.8) in Southeast Asia to 8,608.63 per 100,000 (95%UI 7,674.07–9,485.19) in high-income Asia Pacific regions. The knee was the most commonly affected joint. High BMI and metabolic risks are the only two GBD risk factors for osteoarthritis. From 1990 to 2021, the age-standardized prevalence, incidence, and YLD attributable to osteoarthritis have been on the rise, with substantial international variations across indicators. Countries with high socio-demographic index (SDI) bear a disproportionately high burden of OA, and inequalities in the burden of disease due to differences in SDI between countries have been increasing over time. Conclusions As a major public health problem, the overall global burden of OA has shown an upward trend from 1990 to 2019, including an increase in the number of cases and inequalities in distribution across the globe, which has resulted in significant health losses and economic burdens. In addition, SDI-related inequalities between countries are increasing. In this regard, national public health authorities and the World Health Organization (WHO) should work together to improve diagnosis and early treatment rates by strengthening disease awareness and education, as well as strengthening international cooperation, providing necessary medical assistance to less developed regions, and actively exploring new strategies for the prevention and treatment of OA.