Browse Articles

Transcatheter Repair Versus Surgery for Atrial Versus Ventricular Functional Mitral Regurgitation: A Post Hoc Analysis of the MATTERHORN Trial

LFD-YOLO: a lightweight fall detection network with enhanced feature extraction and fusion

Abstract Falls are one of the significant safety hazards for the elderly. Current object detection models for fall detection often suffer from high computational complexity, limiting their deployment on resource-constrained edge devices. Although lightweight models can reduce computational requirements, they typically compromise detection accuracy. To address these challenges, and considering the more lightweight architecture of YOLOv5 compared to other YOLO series models such as YOLOv8, we propose a lightweight fall detection model based on YOLOv5, named Lightweight Fall Detection YOLO (LFD-YOLO). Our method introduces a novel lightweight feature extraction module, Cross Split RepGhost (CSRG), which reduces information loss during feature map transmission. We also integrate an Efficient Multi-scale Attention (EMA) to enhance focus on the human pose. Moreover, we propose a Weighted Fusion Pyramid Network (WFPN) and utilize Group Shuffle Convolutions (GSConv) to reduce the model’s computational complexity and improve the efficiency of multi-scale feature fusion. Additionally, we design an Inner Weighted Intersection over Union (Inner-WIoU) loss to accelerate model convergence and enhance generalization. We construct a Person Fall Detection Dataset (PFDD) dataset covering diverse scenarios. Experimental results on the PFDD and the publicly available Falling Posture Image Dataset (FPID) datasets show that, compared to YOLOv5s, LFD-YOLO improves mAP0.5 by 1.5% and 1.7%, respectively, while reducing the number of parameters and calculations by 19.2% and 21.3%. Furthermore, compared to YOLOv8s, LFD-YOLO reduces the number of parameters and calculations by 48.6% and 56.1%, respectively, while improving mAP0.5 by 0.3% and 0.5%. These results demonstrate that LFD-YOLO achieves higher detection accuracy and lower computational complexity, making it well-suited for fall detection tasks.

Publisher Correction: Genetic diversity and origin of Kazakh Tobet Dogs

Turbulence enhances wave attenuation of seagrass in combined wave–current flows

The wave attenuation properties of seagrasses are key to accurately predict how effective these plants are at protecting coasts from erosion and floods. While recent studies have significantly advanced the understanding of seagrass wave attenuation in pure-wave conditions, the presence of a current introduces several complications that have yet to be fully explored. In the present study, we quantify the wave attenuation of seagrass canopies in the presence of a current parallel to the direction of wave propagation via experiments conducted with dynamically scaled mimics of seagrass installed in a laboratory flume facility. The dataset we present is the largest of its kind and spans a broad range of wave properties, current velocities, water depths, and plant densities for a total of over 300 experiments. Using our experimental results, we show that the commonly employed approach of modeling wave attenuation as a result of vegetation drag works well for a range of conditions but underpredicts systematically when turbulence generated by the interaction between the seagrass canopy and the current is sufficiently strong. We then employ phenomenological arguments and experimental data to identify a nondimensional parameter that effectively quantifies the relative importance of turbulence and drag in dictating the overall observed wave attenuation. Moreover, we propose a simple but physically based modeling approach that is consistent with the proposed phenomenology and can be used for applications in coastal waters.

Study of the degradation and microstructural characteristics of granite porphyry with freeze–thaw cycles

Thermal-solutal-induced bistability of evaporating multicomponent liquid thin films

Volatile multicomponent liquid films show rich dynamics, due to the complex interplay of gradients in temperature and in solute concentrations. Here, we study the evaporation dynamics of a tricomponent liquid film, consisting of water, ethanol, and trans-anethole oil (known as “ouzo”). With the preferential evaporation of ethanol, cellular convective structures are observed both in the thermal patterns and in the nucleated oil droplet patterns. However, the feature sizes of these two patterns can differ, indicating dual instability mechanisms dominated by either temperature or solute concentration. Using numerical simulations, we quantitatively compare the contributions of temperature and solute concentration on the surface tension. Our results reveal that the thermal Marangoni effect predominates at the initial evaporation stage, resulting in cellular patterns in thermal images, while the solutal Marangoni effect gradually becomes dominant. By regulating the transition time of this thermal-solutal-induced bistability and the nucleation time of oil microdroplets in the ternary mixture, the oil droplet patterns can be well controlled. This capability not only enhances our understanding of the evaporation dynamics but also paves the way for precise manipulation of nucleation and deposition processes at larger scales.

Deep attention model for arrhythmia signal classification based on multi-objective crayfish optimization algorithmic variational mode decomposition

Anomalous suppression of large-scale density fluctuations in classical and quantum spin liquids

Classical spin liquids (CSLs) are intriguing states of matter that do not exhibit long-range magnetic order and are characterized by an extensive ground-state degeneracy. Adding quantum fluctuations, which induce dynamics between these different classical ground states, can give rise to quantum spin liquids (QSLs). QSLs are highly entangled quantum phases of matter characterized by fascinating emergent properties, such as fractionalized excitations and topological order. One such exotic quantum liquid is the Z 2 QSL, which can be regarded as a resonating valence bond (RVB) state formed from superpositions of dimer coverings of an underlying lattice. In this work, we unveil a hidden large-scale structural property of archetypal CSLs and QSLs known as hyperuniformity, i.e., normalized infinite-wavelength density fluctuations are completely suppressed in these systems. In particular, we first demonstrate that classical ensembles of close-packed dimers and their corresponding quantum RVB states are perfectly hyperuniform in general. Subsequently, we focus on a ruby-lattice spin liquid that was recently realized in a Rydberg-atom quantum simulator, and show that the QSL remains effectively hyperuniform even in the presence of a finite density of spinon and vison excitations, as long as the dimer constraint is still largely preserved. Moreover, we demonstrate that metrics based on the framework of hyperuniformity can be used to distinguish the QSL from other proximate quantum phases. These metrics can help identify potential QSL candidates, which can then be further analyzed using more advanced, computationally intensive quantum numerics to confirm their status as true QSLs.

Multifactor prediction model for stock market analysis based on deep learning techniques

Aqueous power source integrated on a microfluidic chip

The growing demand for portable sensors for point-of-care (POC) and onsite health monitoring has led to significant interest in developing suitable power sources. In this study, we developed a microfluidic chip-integrated reverse electrodialysis (μRED) system for ecofriendly power generation with monolithic operation. Leveraging its fully ionic characteristic, μRED was successfully applied to an ionic diode, thereby demonstrating its capability for seamless integration. The feasibility of operating a bipolar electrode sensor without an external power supply was demonstrated, highlighting its broad applicability in electrochemical portable sensors. μRED has great potential for future applications, including electrochemical sensors for POC diagnostics and wearable devices.

An optimized detection model for micro-terrain around transmission lines

Abstract Detecting micro-terrain is essential for the effective layout and maintenance of transmission lines. To address the issues of detection incompleteness, classification ambiguity, and inefficiency in traditional methods, particularly the challenge of distinguishing between saddle and canyon micro-terrain, this paper optimizes the calculation of micro-terrain features and the strategy of micro-terrain detection, and explores a detection method of micro-terrain around transmission lines based on the GPU parallel random forest. This paper employs the GPU parallel random forest model as the extraction framework, leveraging the computational speed advantage of GPU parallel technology for handling large datasets and the robustness inherent in the ensemble approach of random forests. The DEM data of 49 transmission lines in the study area was used for micro-terrain detection experiments. Most of these 49 routes are situated in mountainous regions with complex terrain and contain diverse micro-terrain categories along their paths, rendering them highly representative. The experimental results demonstrate that the proposed method effectively identifies atypical micro-terrain types and four typical micro-terrain types—saddle, canyon, alpine watershed, and uplift—with a classification accuracy of 97.96% and a Kappa coefficient of 0.974. Compared to the traditional method, which achieves a classification accuracy of 75.19% and a Kappa coefficient of 0.642, the proposed method demonstrates a clear improvement in performance. Moreover, by employing the parallel model, the acceleration ratios for training and classification reach 50.57 and 109.06, respectively, significantly improving the efficiency of micro-terrain detection for large-scale regions. These findings could significantly enhance transmission line maintenance and layout planning by providing more accurate micro-terrain data, enabling better decision-making and resource allocation for infrastructure development and disaster risk mitigation.

The role of dietary inflammation in the risk of osteoporosis in Iranian postmenopausal women: a case-control study

Out-of-distribution generalization via composition: A lens through induction heads in Transformers

Large language models (LLMs) such as GPT-4 sometimes appear to be creative, solving novel tasks often with a few demonstrations in the prompt. These tasks require the models to generalize on distributions different from those from training data—which is known as out-of-distribution (OOD) generalization. Despite the tremendous success of LLMs, how they approach OOD generalization remains an open and underexplored question. We examine OOD generalization in settings where instances are generated according to hidden rules, including in-context learning with symbolic reasoning. Models are required to infer the hidden rules behind input prompts without any fine-tuning. We empirically examined the training dynamics of Transformers on a synthetic example and conducted extensive experiments on a variety of pretrained LLMs, focusing on a type of component known as induction heads. We found that OOD generalization and composition are tied together—models can learn rules by composing two self-attention layers, thereby achieving OOD generalization. Furthermore, a shared latent subspace in the embedding (or feature) space acts as a bridge for composition by aligning early layers and later layers, which we refer to as the common bridge representation hypothesis.

Cluster analysis of physical activity and physical fitness and their associations with components of school skills in children aged 8–9 years

The genomic and epigenomic landscapes of hemizygous genes across crops with contrasting reproductive systems

Hemizygous genes, which are present on only one of the two homologous chromosomes of diploid organisms, have been mainly studied in the context of sex chromosomes and sex-linked genes. However, these genes can also occur on the autosomes of diploid plants due to structural variants (SVs), such as a deletion/insertion of one allele, and this phenomenon largely unexplored in plants. Here, we investigated the genomic and epigenomic landscapes of hemizygous genes across 22 genomes with varying propagation histories: eleven clonal lineages, seven outcrossed samples, and four inbred and putatively homozygous genomes. We identified SVs leading to genic hemizygosity. As expected, very few genes (0.01 to 1.2%) were hemizygous in the homozygous genomes, representing negative controls. Hemizygosity was appreciable among outcrossed lineages, averaging 8.7% of genes, but consistently elevated for the clonal samples at 13.8% genes, likely reflecting heterozygous SV accumulation during clonal propagation. Compared to diploid genes, hemizygous genes were more often situated in centromeric than telomeric regions and experienced weaker purifying selection. They also had reduced levels of expression, averaging ~20% of the expression levels of diploid genes, violating the evolutionary model of dosage compensation. We also detected higher DNA methylation levels in hemizygous genes and transposable elements, which may contribute to their reduced expression. Finally, expression profiles showed that hemizygous genes were more specifically expressed in contexts related to fruit development, organ differentiation, and stress responses. Overall, hemizygous genes accumulate in clonally propagated lineages and display distinct genetic and epigenetic features compared to diploid genes, shedding unique insights into genetic studies and breeding programs of clonal crops.

Investigating brain activity at rest in patients with persistent genital arousal disorder (PGAD) using functional magnetic resonance imaging

Abstract Persistent genital arousal disorder (PGAD) is a rare disease causing high emotional distress eminently impacting the individual’s quality of life. Experts in this field assume that the disease is caused by a multifaceted interplay of different etiologies which may share a common neurobiological basis. However, only one functional neuroimaging investigation exist, and a more in-depth comprehension of the neurobiological foundation is required. Therefore, this study aims to provide new insights into how the functional integration of brain regions may relate to PGAD. By using the functional magnetic resonance imaging (fMRI) technique, functional connectivity at rest (rs-FC) was compared between patients suffering PGAD (n = 26) and healthy controls (n = 26). Patients with PGAD showed different pattern in connectivity within brain structures putatively associated with the psychological and somatic dimensions of the disease including the right amygdala, left anterior cingulate cortex, right insula cortex, thalamic nuclei and prefrontal regions as seeds. The majority of these showed differences in brain connectivity pattern to the precuneus and prefrontal regions. The study offers preliminary insights into the characteristics and relevant neural mechanisms of PGAD. Nevertheless, since this study did not identify any peripheral correlates that would corroborate the interpretation of these findings, they were interpreted from a more theoretical perspective, thereby offering potential areas of focus for future research.

Contextual neural dynamics during time perception in the primate ventral premotor cortex

Understanding how time perception adapts to cognitive demands remains a significant challenge. In some contexts, the brain encodes time categorically (as “long” or “short”), while in others, it encodes precise time intervals on a continuous scale. Although the ventral premotor cortex (VPC) is known for its role in complex temporal processes, such as speech, its specific involvement in time estimation remains underexplored. In this study, we investigated how the VPC processes temporal information during a time interval comparison task (TICT) and a time interval categorization task (TCT) in primates. We found a notable heterogeneity in neuronal responses associated with time perception across both tasks. While most neurons responded during time interval presentation, a smaller subset retained this information during the working memory periods. Population-level analysis revealed distinct dynamics between tasks: In the TICT, population activity exhibited a linear and parametric relationship with interval duration, whereas in the TCT, neuronal activity diverged into two distinct dynamics corresponding to the interval categories. During delay periods, these categorical or parametric representations remained consistent within each task context. This contextual shift underscores the VPC’s adaptive role in interval estimation and highlights how temporal representations are modulated by cognitive demands.

Design and fabrication of an ultra small quadband diplexer integrated with a diplexed power amplifier for mid band 5G applications

tRNA selectivity during ribosome-associated quality control regulates the critical sterility-inducing temperature in two-line hybrid rice

The two-line hybrid rice system, a cutting-edge hybrid rice breeding technology, has greatly boosted global food security. In thermo-sensitive genic male sterile (TGMS) lines, the critical sterility-inducing temperature (CSIT; the temperature at which TGMS lines change from male fertile to complete male sterile) acts as a key threshold. We recently uncovered that thermo-sensitive genic male sterility 5 ( tms5 ), a sterile locus presenting in over 95% of TGMS lines, leads to the overaccumulation of 2′,3′-cyclic phosphate (cP)-ΔCCA-tRNAs and a deficiency of mature tRNAs, which underlies the molecular mechanism of tms5 -mediated TGMS. However, there are a few reports on the regulatory mechanism controlling CSIT. Here, we identified a suppressor of tms5 , an amino acid substitution (T552I) in the rice Rqc2 (ribosome-associated quality control 2), increases the CSIT in tms5 lines through its C-terminal alanine and threonine modification (CATylation) activity. This substitution alters tRNA selectivity, leading to the recruitment of different tRNAs to the A-site of ribosome and CATylation rate by OsRqc2 during ribosome-associated quality control (RQC), a process that rescues stalled ribosomes and degrades abnormal nascent chains during translational elongation. Further, the mutation restores the levels of mature tRNA-Ser/Ile to increase the CSIT of tms5 lines. Our findings reveal the origin of overaccumulated cP-ΔCCA-tRNAs in tms5 lines, further deepening our understanding of the regulatory network in governing CSIT of TGMS lines containing tms5.

Health risk assessment via Monte Carlo simulation and sensitivity analysis for fluoride and nitrate content in bottled waters consumed in Kermanshah city, Iran

Abstract Bottled water consumption has increased in recent decade due to many reasons, especially significant decline in water quality and quantity. The concentration of fluoride and nitrate in bottled waters may vary based on brands and locations. This study was carried out to determine the levels of fluoride and nitrate in bottled waters consumed in Kermanshah city and assess the related non -carcinogenic risks. Totally, 22 brands of bottled water were collected from markets. Fluoride and nitrate measurement was conducted via a UV-visible spectrophotometer (DR-5000). From the results, Fluoride and nitrate levels in the studied bottled waters were 0.32 ± 0.18 mg/L and 2.3 ± 1.41 mg/L, respectively. The risk of non-carcinogenic in term of HQ for fluoride exposure, for only 2 brands of bottled water were > 1 for infants group. HQ was less than 1 for nitrate in all the brands for all the age groups revealed non-carcinogen risks. Hazard index (HI) calculation showed that only in 2 brands of bottled water HI was > 1 for infants group. The HI were as infants (0.64) > children (0.36) > teenagers (0.27) > adults (0.24). From Monte Carlo simulation, 95th Percentile for nitrate and fluoride was less than 1 for all the groups. This result indicated non-carcinogenic risks of nitrate and fluoride for 95% of the studied groups. Moreover, sensitivity analysis showered that concentration for both nitrate and fluoride had the highest effect on HQ for all the groups. From this work, although fluoride and nitrate content in the bottled waters were at standard range, but infants were proportionally at higher risk.