Microalgae are promising photosynthetic platforms for high-value compounds, yet their industrial use is often hindered by a trade-off between robust growth and the metabolic burden of payload production or cell wall disruption. Constitutive engineering for these traits compromises cultivator fitness. Here we report the development of a versatile, thermal-regulated "PLUG-IN" chassis in Nannochloropsis oceanica that enables programmable control of metabolic output and cell integrity. Comparative transcriptomics identify two highly heat-inducible promoters (PNoED and PNoUK), which we use to construct a modular thermal gene-amplification. Heat-activated AtWRI1 expression enhances triacylglycerol and eicosapentaenoic acid accumulation, while temperature-dependent silencing of the cellulose synthase gene CesA1 triggers rapid cell-wall weakening without affecting growth under permissive conditions. Coupling metabolic and cell-wall modules yields strains capable of grind-free lipid recovery and substantially improved intracellular product accessibility. Notably, these engineered thermally controlled programs remain functional in mammalian hosts, demonstrating cross-kingdom compatibility. This work establishes a "plug-in" chassis compatible with mammalian systems that synchronizes growth, production, and cell-wall re-configuration, providing a versatile platform for photosynthetic bioproduction and microalgal synthetic biology.
In vivo gene therapy is rapidly advancing to treat patients with a wide range of genetic disorders. Recombinant adeno-associated virus (rAAV) vectors provide long-term gene expression, low immunogenicity, and adaptable tissue tropism. Production of rAAV in insect cells with the Baculovirus Expression Vector System currently depends on infection with multiple baculoviruses encoding AAV genes Rep, Cap, and the therapeutic gene. In this research article, we present simplified rAAV production using a single baculovirus vector with improved genetic stability, called BAC6Rep, with AAV Rep locked into the baculovirus genome. Robust AAV Rep and Cap expression was maintained upon serial passaging, and high rAAV yields (up to 3.5e11 genome copies/ml) were obtained at high and low multiplicities of infection. Next-generation sequencing revealed a significantly improved genetic stability of BAC6Rep, which can now be used as a plug-in hybrid vector for flexible insertion of different AAV Cap and gene of interest tailored to the therapeutic target.
Lipid accumulation in microalgae is typically induced by stress but often comes at the expense of growth and energy stability. Here, we introduced the plsXYC module from Cyanobacterium aponinum into Chlamydomonas reinhardtii, aiming to enhance stress-responsive lipid production by a plug-in prokaryotic acyltransferase route. The initial wild-type strain produced lipids at 135.1 mg/g-dry cell, while the engineered CrXYC sustained the highest lipid yield of 385 mg/g-DCW across light and dark cycles. Transcriptomics revealed that light-dark shifts drove 65% of expression variance and elevated ATP levels. During heterotrophic culture under nitrogen starvation, CrXYC preserved ATP up to 1.6-fold higher than the parental background. Carbon repartition was proved using 13C-isotopes, and redox reinforcement showed coordinated upregulation of lipid assembly pathways and repression of starch biosynthesis. This shift coincided with enhanced superoxide scavenging activity, while broad antioxidant capacity remained unchanged. Together, we demonstrate that introducing an orthogonal acyl-acyl carrier protein entry route sustains lipid accumulation under stress while preserving growth and energy balance.
The paper will deal with the problem of optimal charging and discharging coordination of Plug-in Electric Vehicles in a local multi-energy system that will be conducted under uncertain conditions. The current practices are usually based on simplified assumptions and fail to coordinate various energy sources and resources effectively, which restricts their usage in real-life conditions. In order to eliminate these constraints, a hybrid optimization model that involves Rider Optimization and Osprey Optimization integration is suggested. The model developed will take into account both the economic and environmental goals, such as the cost of operation, profitability of the system, and minimization of carbon emission. The lack of certainty in the behavior and generation of the renewable is factored in by using probabilistic modeling, which allows the system to be more realistic. The suggested algorithm constraints global exploration and local exploitation to optimally search the energy scheduling strategies. The outcomes of the simulation display the assumption that the proposed method can deliver the enhanced performance of the operational level as it successfully organizes the activities of distributed energy resources and electric vehicle charging. The model offers computationally effective and flexible framework to be used in practical energy management applications in local multi-energy systems.
This study compared two widely used biomechanical models-Plug-in Gait (PiG) and Conventional Gait Model 2.3 (CGM2.3)-during overground walking (WALK) and single-leg squats (SLS) in 24 healthy adults. Data was collected using a 20-camera Vicon system and force plates. Static trials were analyzed with medial knee and ankle markers to align joint axes across models. Kinematic and kinetic outputs were compared using root mean square differences (RMSD) and statistical parametric mapping (SPM) paired t-tests. During WALK, PiG produced greater internal rotation at the knee (RMSD 17.8°, p < 0.001) and hip (RMSD 5.0°, p < 0.001), and smaller sagittal-plane flexion angles (RMSD 2.6° knee, 2.3° hip) compared with CGM2.3. In single-leg squats, these discrepancies increased to 29.1° and 9.0°, respectively, with sagittal-plane differences of 4.4° at the knee and 5.1° at the hip. CGM2.3 yielded higher knee flexion moments (31% in WALK, 104% in SLS), while PiG produced higher frontal-plane knee moments (28% and 89%). The differences were most pronounced at deeper flexion angles. These results demonstrate that biomechanical outcomes differ systematically between models, emphasizing the impact of model selection on joint kinematics and kinetics in human movement analysis.
Accurate and efficient weed identification is essential for precision agriculture, yet existing deep learning approaches struggle to balance classification accuracy with computational efficiency required for real-time field deployment. Fine-grained weed classification presents unique challenges due to subtle inter-species morphological differences and significant intra-class variations caused by diverse growth stages and environmental conditions. In this paper, we propose a unified RepVGG model factory with plug-in Squeeze-and-Excitation (SE) attention and Generalized Mean (GeM) pooling for fine-grained agricultural classification. Our approach leverages the structural re-parameterization technique of RepVGG to achieve efficient inference while enhancing discriminative feature learning through channel-wise attention recalibration and adaptive feature aggregation. The SE attention module explicitly models inter-channel dependencies to emphasize informative features, while the learnable GeM pooling enables adaptive interpolation between average and max pooling behaviors for optimal feature aggregation. We further develop a unified model construction interface that supports systematic benchmarking across multiple deep learning backends with consistent experimental protocols. Comprehensive experiments on the CottonWeedID15 dataset demonstrate that our proposed RepVGG-B1 + SE + GeM model achieves state-of-the-art performance with a testing F1-score of 99.5%, outperforming 27 baseline architectures including ResNet101 (99.1%), DenseNet161 (98.9%), and EfficientNet variants. Notably, our model maintains superior inference efficiency at 188.7 ms, which is 8.8% faster than ResNet101 while achieving 0.4% higher accuracy. Ablation studies confirm that SE attention consistently improves accuracy by 0.3% across different RepVGG backbones with minimal parameter overhead (2-3%), while GeM pooling provides complementary gains of 0.1-0.2%. The proposed unified framework offers a flexible and modular solution for deploying high-performance weed identification systems in resource-constrained agricultural environments.
The labor- and experience-intensive creation of 3D assets with physically based rendering (PBR) materials demands an autonomous 3D asset creation pipeline. However, most existing 3D generation methods focus on geometry modeling, either baking textures into simple vertex colors or leaving texture synthesis to post-processing with image diffusion models. To achieve end-to-end PBR-ready 3D asset generation, we present Lightweight Gaussian Asset Adapter (LGAA), a novel framework that unifies the modeling of geometry and PBR materials by exploiting multi-view (MV) diffusion priors from a novel perspective. The LGAA features a modular design with three components. Specifically, the LGAA Wrapper reuses and adapts network layers from MV diffusion models, which encapsulate knowledge acquired from billions of images, enabling better convergence in a data-efficient manner. To incorporate multiple diffusion priors for geometry and PBR synthesis, the LGAA Switcher aligns multiple LGAA Wrapper layers encapsulating different knowledge. Then, a tamed variational autoencoder (VAE), termed LGAA Decoder, is designed to predict 2D Gaussian Splatting (2DGS) with PBR channels. Finally, we introduce a dedicated post-processing procedure to effectively extract high-quality, relightable mesh assets from the resulting 2DGS. Extensive quantitative and qualitative experiments demonstrate the superior performance of LGAA with both text- and image-conditioned MV diffusion models. Additionally, the modular design enables flexible incorporation of multiple diffusion priors, and the knowledge-preserving scheme effectively preseves the 2D priors learned on massive image dataset, which leads to data efficient finetuning to lift the MV diffuison models for 3D generation with merely 69 k multi-view instances.
In randomized clinical trials (RCTs), the accurate estimation of marginal treatment effects is crucial for determining the efficacy of interventions. Enhancing the statistical power of these analyses is a key objective for statisticians. The increasing availability of historical data from registries, prior trials, and health records presents an opportunity to improve trial efficiency. However, many methods for historical borrowing compromise strict type I error rate control. Building on the work by Schuler et al. on prognostic score adjustment for linear models, this paper extends the methodology to the plug-in analysis proposed by Rosenblum and van der Laan using generalized linear models (GLMs) to further enhance the efficiency of RCT analyses without introducing bias. Specifically, we train a prognostic model on historical control data and incorporate the resulting prognostic scores as covariates in the plug-in GLM analysis of the trial data. This approach leverages the predictive power of historical data to improve the precision of marginal treatment effect estimates. We demonstrate that this method achieves local semi-parametric efficiency under the assumption of an additive treatment effect on the link scale. We expand the GLM plug-in method to include Negative Binomial regression. Additionally, we provide a straightforward formula for conservatively estimating the asymptotic variance, facilitating power calculations that reflect these efficiency gains. Our simulation study supports the theory. Even without an additive treatment effect, we observe increased power or reduced standard error. While population shifts from historical to trial data may dilute benefits, they do not introduce bias. We demonstrate reductions in the estimated variance through an analysis of clinical trial data provided by Novo Nordisk A/S.
BACKGROUND: Traditional methods for producing custom-made orthoses are often time-consuming, labor-intensive, and reliant on manual processes, which limit both scalability and the degree of individualization. The development of 3D scanning technologies, computer-aided design (CAD), and additive manufacturing offers a promising alternative enabling patient-specific solutions with greater precision, speed, and efficiency. This study aimed to create an algorithm for automating the design process of personalized knee orthoses based on 3D scanning and intended for 3D printing production. METHODS: A parametric modeling workflow was developed in the Rhino environment using the Grasshopper plug-in to streamline personalized knee orthoses creation. The process began with acquiring high-quality 3D scans using Structure Sensor Mark II scanner mounted on an iPad with 3DsizeMe software. The parametric algorithm was transformed into an autonomous Rhino plug-in using C# language and RhinoCommon API. As part of Post-Market Clinical Follow-up (PMCF), three participants with knee joint disorders used orthoses for one month. Assessment used a 5-point scale (1 = poor, 5 = excellent). Personalized orthoses were manufactured using powder-bed fusion technology with PA11 CF nylon powder reinforced with carbon fibers. RESULTS: Design time was reduced from approximately 8 h to 10,3 ± 1,4 min. In Grasshopper prototype phase, average design time was 26,7 ± 4,5 min. Following the implementation of the Rhino plug-in, the design time was further reduced to approximately 10 min. The tool was shown to meet user requirements and fulfill its intended purpose. All three PMCF participants rated orthoses positively, reporting high comfort, effective stabilization, increased physical activity, and overall satisfaction with functionality and appearance. Participants P1 and P2 noted a large increase in physical activity, with P1 indicating pain reduction that increased mobility. CONCLUSIONS: This study demonstrates that the combined use of Rhino and Grasshopper provides an effective platform for parametric design of personalized knee orthoses based on patient-specific 3D scans. The workflow reduced design time to approximately 10,3 ± 1,4 min, highlighting potential for routine clinical applications. This reduction is economically significant, lowering labor costs and implementation thresholds for personalized orthotic solutions in clinical practice.
Evaluating intrusion detection methods at the level of individual MITRE Adversarial Tactics, Techniques, and Common Knowledge (ATT&CK) for Industrial Control System techniques requires Operational Technology traffic in which each attack sequence carries its MITRE technique identifier as ground truth. Publicly available Industrial Control System datasets either provide coarse attack-versus-benign labels (SWaT, WADI, CIC-APT-IIoT) or require ex-post technique reconstruction from CALDERA operation logs, and therefore do not support per-technique benchmarking. We describe one primary contribution and two supporting contributions, demonstrated on one Modbus/Raspberry-Pi programmable logic controller/CALDERA/convolutional bidirectional Long Short-Term Memory autoencoder (CNN-BiLSTM-AE) use case. The primary contribution is an in-orchestrator labelling methodology for per-technique-labelled Industrial Control System attack capture. Its single load-bearing property is that the campaign orchestrator owns the label primitive and writes each per-sequence technique identifier into the capture artefact at injection time, eliminating ex-post log-to-packet alignment. The first supporting contribution is a protocol-aware detection pipeline. Its load-bearing architectural choice is a priority-ordered protocol router that dispatches each labelled flow to a per-protocol detector plug-in (protocol-aware features here, with generic-flow features admissible as an alternative plug-in policy on the same router). The second supporting contribution is a suite of four reproducible CALDERA chains (three Information-Technology-to-Operational-Technology kill chains plus one enterprise-side control) that exercise the labelling methodology end-to-end and the detection pipeline along complementary detection paths. All three contributions are platform-independent: any ATT&CK-aligned emulator and any fieldbus protocol can host the labelling methodology, and any detector trained on an admissible feature space can plug into the router. The dataset contains 40,000 benign and 9997 attack Modbus sequences spanning four ATT&CK techniques (T0802 Automated Collection, T0831 Manipulation of Control, T0836 Modify Parameter, T0846 Remote System Discovery). On this dataset, the CNN-BiLSTM-AE reaches a 100% true-positive rate (TPR) at the 98th-percentile benign threshold across all four techniques and a 99.7% overall TPR at the tighter 99.5th-percentile threshold, with per-technique TPR between 96.1% (T0836 Modify Parameter) and 100% (T0802 Automated Collection, T0846 Remote System Discovery). Across the four CALDERA chains, the Modbus autoencoder produces 234 protocol-layer detections and the Security Information and Event Management (SIEM) rule set produces 30 alerts, with per-chain tactic coverage between 0.714 and 0.786 and CALDERA-ability success rates between 0.800 and 0.857.
This study aims to investigate the prevalence and influencing factors of dry eye among Chinese nurses, and to explore the relationship between nurses' perceived stress, sleep quality and dry eye. This study adopted a cross-sectional survey design and distributed electronic questionnaires through Wechat platform. A total of 450 valid questionnaires from nurses across the country were collected. The survey included demographic and sociological information on nurses, dry eye symptoms (Ocular Surface Disease Index), sleep quality (Pittsburgh Sleep Quality Index), and perceived stress (Perceived Stress Scale). Generalized linear model was used to analyze the factors affecting the symptoms of dry eye in nurses. The Process plug-in in SPSS was used to conduct mediation analysis using Bootstrap method. Among the 450 nurses included, the prevalence of dry eye reached 66%. Specifically, mild dry eye accounted for 27.56%, moderate dry eye accounted for 14.89%, severe dry eye accounted for 23.56%. Multivariate analysis revealed that years in nursing, hospital grade, contact lens and frame glasses wearing, sleep quality and perceived stress were influencing factors for dry eye. Mediation analysis revealed that the bootstrap 95% confidence interval for the mediating effect of PSS score on OSDI score via PSQI score does not include zero. The prevalence of dry eye in Chinese nurses is very high, and the proportion of moderate to severe dry eye is relatively large. Years in nursing and wearing contact lenses or rimmed glasses were risk factors for dry eye in nurses. The perceived pressure and sleep quality are the direct risk factors of dry eye, and the perceived pressure can also influence the occurrence and development of dry eye through the mediating effect of sleep quality.
Non-suicidal self-injury (NSSI) is a common but maladaptive behavior among adolescents. Previous studies demonstrated a significant association of anxiety and depression with NSSI. However, the psychological and neuroendocrine mechanisms through which negative emotions influence NSSI remain unclear. The present study aimed to investigate whether the personality trait of neuroticism mediates the relationship between negative emotions and NSSI, and whether thyroid hormones moderate this pathway. A total of 104 Han Chinese adolescents and young adults (aged 12-22 years) who exhibited NSSI behaviors were recruited. The participants completed questionnaires to assess NSSI severity (Questionnaire for Middle School Students' Behavior, QMSSB), depressive symptoms (Hamilton Depression Rating Scale, HAMD), anxiety symptoms (Hamilton Anxiety Rating Scale, HAMA) and personality traits (Eysenck Personality Questionnaire, EPQ). Venous blood was collected for thyroid function tests (TT3, TT4, FT3, FT4 and TSH). Mediation and moderation analyses were conducted using the PROCESS plug-in for SPSS software. Neuroticism mediated the relationship between emotional symptoms and NSSI. For HAMA in particular, neuroticism fully mediated the association with NSSI (effect = 0.0798, 95% bootstrap confidence interval [0.0387, 0.1367]). TT3 (b = 0.3112, p = 0.0147) and FT3 (b = 0.1118, p = 0.0371) positively moderated the HAMD-NSSI relationship, while TSH negatively moderated this relationship in the full sample (b = -0.0976, p = 0.0011) and remained significant in the high-neuroticism subgroup (β = -1.1116, p = 0.0026). Following FDR correction, TT3 (q = 0.0245), FT3 (q = 0.0412) and TSH (q = 0.0055) were found to significantly moderate the HAMD-NSSI relationship in the entire sample. The negative moderating effect of TSH remained significant in the high-neuroticism subgroup (q = 0.0078). However, no moderating effects in the low-neuroticism subgroup survived FDR correction (all q > 0.05). No significant moderating effects of thyroid hormones were found on the HAMA-NSSI pathway (all p > 0.05). Neuroticism mediates the mood-NSSI link. TT3 and FT3 positively moderate the depression-NSSI link, while TSH negatively moderates it. This negative moderating effect of TSH remains significant only in the high-neuroticism subgroup after FDR correction. These findings integrate psychological and neuroendocrine mechanisms for the identification and intervention of NSSI risk in youths.
Electrified vehicles can substantially reduce emissions from light-duty vehicles (LDVs), but large-scale deployment remains challenging due to their associated demands for critical materials in batteries. The challenge is further complicated by trade-offs among greenhouse gas emissions, costs, and critical materials. We develop an optimization model to explore the cost and technical feasibility of meeting climate targets for U.S. LDVs under various material supply scenarios. To meet a sectoral target consistent with 2 °C, global lithium supply would need to grow by 35%/yr to 2035 or 50%/yr to 2030, assuming: (1) no recycling, (2) the U.S. can access a share of global supply proportionate to its population, and (3) medium- and heavy-duty vehicles electrify as fast as LDVs. Recycling or greater U.S. material allocation (proportionate to gross domestic product) can reduce the necessary growth rates to 30-45%/yr or 5-10%/yr, respectively. In low material supply scenarios, the 2 °C target is sometimes still attainable with preference to hybrid vehicles in the early years, later transitioning to a mix of fully electric and plug-in hybrid vehicles (PHEVs) from the 2030s onward. To hedge against future material supply uncertainty, PHEVs can act as transitional technologies in the short term and remain an important technology in the long term.
This article investigates the dissemination of geometric figures for education and academic research in architecture. When extracting 3D geometry and metadata from digital models created using computer-aided design (CAD) software, traditional interoperability workflows based on file exchange often introduce lossy conversions to standardized data schemas. This paper proposes an alternative approach based on a direct data stream between CAD software and a web-based visualization environment, using the open-source collaborative platform Speckle. After outlining the objectives and methods, a state-of-the-art review highlights key technical challenges. A case study based on geometric figures modeled in TopSolid is then presented, involving two development layers: (1) a Speckle connector implemented as a plug-in within TopSolid, and (2) a customized web viewer built using speckle-viewer and Three.js. The results show that a CAD environment can be effectively replicated on the web, enabling flexible and open dissemination of academic content in architectural geometry.
Causal effects are often characterized with population summaries. These might provide an incomplete picture when there are heterogeneous treatment effects across subgroups. Since the subgroup structure is typically unknown, it is more challenging to identify and evaluate subgroup effects than population effects. We propose a new solution to this problem: Causal k-Means Clustering, which leverages the k-means clustering algorithm to uncover the unknown subgroup structure. Our problem differs significantly from the conventional clustering setup since the variables to be clustered are unknown counterfactual functions. We present a plug-in estimator which is simple and readily implementable using off-the-shelf algorithms, and study its rate of convergence. We also develop a new bias-corrected estimator based on nonparametric efficiency theory and double machine learning, and show that this estimator achieves fast root-n rates and asymptotic normality in large nonparametric models. Our proposed methods are especially useful for modern outcome-wide studies with multiple treatment levels. Further, our framework is extensible to clustering with generic pseudo-outcomes, such as partially observed outcomes or otherwise unknown functions. Finally, we explore finite sample properties via simulation, and illustrate the proposed methods using a study of mobile-supported self-management for chronic low back pain.
The severity of chronic kidney disease (CKD) stage 5D complications increases with hemodialysis (HD) vintage, leading to elevated mortality risk. Calibrating the cumulative risk of HD patients against the baseline aging rate in the general population enables a quantitative assessment of dialysis-related accelerated aging. To develop the NephroAge framework for quantifying accelerated aging in patients on maintenance HD. The study included 5356 patients aged ≥ 40 years initiating first-time maintenance HD. Baseline hazard function was estimated using the Human Mortality Database life tables. Age-standardized expected survival was calculated using a plug-in estimator conditional on age at HD initiation. The cumulative incidence of death was estimated using the Aalen-Johansen estimator. For each patient, we calculated an equivalent age in the general population at which the cumulative hazard matched that observed on HD. NephroAge was defined as age at HD start plus HD vintage plus Δ, where Δ represents dialysis-attributable "added years" of aging. Added years showed a strongly nonlinear pattern with both HD vintage and age at initiation, reflecting the nonadditivity of their effects. Patients who started HD at very advanced ages (75-80 years) had NephroAge values close to or below their chronological age, consistent with strong survival selection in this highly resilient subgroup. NephroAge condenses age at HD initiation and treatment vintage into a single, interpretable metric of premature aging. Younger patients accumulate substantially more added years per year on HD than their older counterparts, highlighting pronounced "biological cost" of long-term dialysis in younger individuals.
To explore the psychological mechanism of perceived stress on achievement motivation (including motivation to pursue success and avoid failure) in medical college students, and to verify the partial mediating effect of psychological resilience in this relationship. A total of 2,691 medical college students in Jining were investigated by random whole cluster sampling, and assessed with Achievement Motivation Inventory (AMS), Chinese Perceived Stress Scale (CPSS) and Connor-Davidson Psychological Resilience Scale (CD-RISC-10). The mediating effect was tested by Model IV of PROCESS plug-in in SPSS 27.0. (1) Achievement motivation had no significant difference in demographic characteristics; psychological stress had significant differences in grade, gender, major preference and family annual income (p < 0.05); psychological resilience had significant differences in all demographic characteristics (p < 0.05). (2) Achievement motivation was significantly negatively correlated with psychological stress (r = -0.419, p < 0.001), and positively correlated with psychological resilience (r = 0.492, p < 0.001); psychological stress was significantly negatively correlated with psychological resilience (r = -0.655, p < 0.001). (3) Psychological resilience played a partial mediating role between stress and achievement motivation: the mediating effect accounted for 59.52% of the total effect on motivation to pursue success and 17.24% on motivation to avoid failure. Perceived stress is directly negatively associated with college students' achievement motivation, and is also indirectly associated with it via the reduction of psychological resilience. Enhancing psychological resilience may buffer the negative association between stress and achievement motivation.
BackgroundIncreasing evidence shows that, compared with the general population, the prevalence of coronavirus disease 2019 is significantly higher in patients with psoriasis treated with immunosuppressive therapy. However, the underlying mechanisms have not yet been clarified.Materials and methodsThe aim of this study was to further investigate the molecular mechanisms underlying these two diseases. Gene expression profiles for coronavirus disease 2019 and psoriasis were downloaded from the gene expression omnibus database (GSE150316 and GSE30999). After identifying the common differentially expressed genes between coronavirus disease 2019 and psoriasis, functional annotation, protein-protein interaction network analysis, module construction, and hub gene identification were performed. Finally, transcription factor-gene regulatory and transcription factor-miRNA regulatory networks of the hub genes were constructed.ResultsA total of 306 common differentially expressed genes, including 168 upregulated genes and 138 downregulated genes, were identified and used for subsequent analysis. According to Kyoto Encyclopedia of Genes and Genomes enrichment analysis, the Rap1 signaling pathway, axon guidance, and focal adhesion contribute to the occurrence and development of coronavirus disease 2019 and psoriasis. Finally, five hub genes, namely EGF, IL1B, SERPINA1, CD8A, and WNT5A, were identified using the CytoHubba plug-in. Among them, three hub genes (EGF, SERPINA1, and CD8A) showed good diagnostic marker value for coronavirus disease 2019 and psoriasis.ConclusionOur findings suggest that coronavirus disease 2019 and psoriasis share common molecular mechanisms that are driven by several specific hub genes. This study provides new insights into the relationship between coronavirus disease 2019 and psoriasis.
Foundation models for embodied artificial intelligence (Embodied AI) increasingly adopt diffusion modules as the action generation core of vision-language-action (VLA) policies, but the diffusion module's iterative denoising imposes prohibitive inference latency for real-time deployment. We address this bottleneck in isolation by rethinking the diffusion action generation module itself. We present Fast Robot Motion Diffusion (FRMD) , a fast robot motion diffusion framework that (i) operates in trajectory-parameter space by predicting movement-primitive coefficients in a low-dimensional manifold, and (ii) collapses multi-step sampling into a single inference step via trajectory-level consistency distillation over the probability-flow ordinary differential equation (ODE). Concretely, FRMD replaces stepwise action generation with a one-pass mapping from noise to full trajectories, followed by a fixed-cost basis expansion; this reduces policy latency from hundreds to tens of milliseconds without modifying upstream vision or language encoders. On standard robotic manipulation task benchmarks, FRMD attains 7 times faster than the vanilla diffusion policy and 10 times faster than the state-of-the-art MPD method, while matching the task success of multi-step diffusion policies. By targeting the diffusion component used throughout VLA systems, FRMD provides a plug-in, latency-optimized motion generator that preserves the advantages of diffusion and makes real-time embodied AI feasible.
This research proposes an empirical benchmarking study of an attention-infused deep convolutional framework for multi-label thoracic pathology classification in chest radiographs, designed to emulate how radiologists selectively focus on suspicious regions. Existing CNN models process entire images uniformly, often missing fine-grained or subtle abnormalities that require localized visual emphasis. In spite of the progress of deep convolutional neural networks (CNNs) in automated CXR analysis, traditional architectures do not sufficiently localize fine-grained pathological information, especially in multi-label contexts. To overcome these constraints, we introduce an attention-aware deep convolutional paradigm that can easily add lightweight spatial attention modules to multiple high-performance CNN backbones, including ResNet101, EfficientNet-B0/B3, and MobileNetV2. The proposed spatial attention module specifically targets the challenge of spatial feature reweighting to improve localization of fine-grained pathological regions; it does not explicitly model other inherent challenges of multi-label CXR classification such as label noise, class imbalance, or disease co-occurrence, which remain contextual factors of the task. The spatial attention system is based on recreating the radiologist attention mechanism, which temporarily accentuates the relevant areas in diagnostics and inhibits the background noise. This improves feature recalibration on intermediate layers, allowing the network to learn global context and localized pathologies simultaneously with little computational cost. It is trained and tested on the NIH ChestX-ray14 dataset with a multi-label classification with sigmoid outputs and binary cross-entropy losses operating on 13 thoracic conditions. It is important to note that Attention MobileNetV2 achieved a micro-AUC improvement of 0.818 to 0.826, whilst Attention ResNet101 achieved its highest micro-AUC of 0.872, which is greater than the non-attention ResNet101 (0.853). Localized pathologies, including emphysema (improvement in performance, +0.05 AUC), effusion (improvement in performance, +0.04) and pneumothorax (improvement in performance, +0.04), showed some of the strongest improvements in performance. ROC analysis showed an improved early-lift behavior and decreased false-positives -which is important in triage. Ablation experiments verified that spatial attention is a selective, architecture-specific promoter, not an amplifier, and the largest advantage was found in lightweight and deep residual networks. Our results demonstrate that spatial attention is a modest but architecture-dependent enhancement of CNN-based chest radiograph classification. The magnitude of improvement varies considerably across backbone architectures: Attention ResNet101 achieves the largest micro-AUC gain (from 0.853 to 0.872), MobileNetV2 shows a moderate improvement (from 0.818 to 0.826), while EfficientNet-B0 and EfficientNet-B3 exhibit minimal change, reflecting diminishing returns in already highly optimized architectures. These results confirm that the proposed spatial attention module functions as a selective, architecture-specific enhancer rather than a universally transformative component. Claims of broad clinical reliability are tempered by the reliance on NLP-derived, noisy labels from ChestX-ray14; this work is best understood as an empirical foundation for future hybrid attention strategies that may combine spatial, channel-wise, and self-attention for more robust multi-label CXR analysis. The proposed framework achieves a practical balance between performance, interpretability, and computational efficiency, and may serve as a principled modular plug-in in future AI-assisted radiological diagnostic pipelines, pending further clinical validation on expert-annotated datasets.