Search Results (954)

Accurate body-weight measurement is important for precision pig farming, but conventional weighing methods are labor-intensive and may disturb normal animal activity. Although three-dimensional sensing systems can provide reliable geometric information, their deployment cost limits large-scale application in commercial farms. This study proposes a non-contact pig weight estimation framework based on standard RGB images. The framework combines SAM 2 foreground extraction with a transformer-based dorsal segmentation network to obtain stable body contours under complex farm conditions. Cross-covariance attention and local patch interaction modules are introduced to preserve both global body structure and local boundary details during segmentation. A hybrid loss function combining focal loss and label-distribution-aware margin loss is further adopted to address foreground-background imbalance. After segmentation, 17 morphometric features are extracted from the dorsal region and used for weight prediction with XGBoost regression. Experiments were conducted on the public PIGRGB-Weight dataset containing 12,476 RGB images from 124 pigs. The proposed method achieved a mean absolute error of 2.983 kg and an $R^2$ value of 0.9891. Compared with a DeepLabV3+-based baseline under the same regression protocol, the proposed framework reduced the prediction error by 24.1%. The results indicate that improving dorsal segmentation quality can substantially enhance the stability of morphometric feature extraction from low-cost RGB images. Full article

Background: Resveratrol is a plant-derived polyphenol with reported anti-inflammatory activity, and the MCP-1/CCL2 axis is a key mediator of monocyte recruitment and inflammatory tissue remodeling. Although individual preclinical studies have examined resveratrol effects on MCP-1/CCL2-related outcomes, the overall in vivo evidence has not been quantitatively synthesized. This systematic review and meta-analysis evaluated whether resveratrol treatment is associated with reduced MCP-1/CCL2-related inflammatory readouts in animal models. Methods: The protocol was registered in PROSPERO (CRD420261339126), and reporting followed the PRISMA 2020 statement. PubMed was searched from inception to 12 March 2026, with additional reference-list screening. Eligible studies were in vivo animal experiments comparing resveratrol-treated and control groups with extractable quantitative MCP-1/CCL2-related outcomes. Effect sizes were calculated as Hedges’ g with 95% confidence intervals and pooled using random-effects models fitted by restricted maximum likelihood. Subgroup, sensitivity, cumulative, influence, funnel-plot, dose meta-regression, and SYRCLE-based risk-of-bias analyses were conducted. Results: Twenty-seven studies contributing 29 analyzable datasets were included. The overall pooled effect was −3.74 (95% confidence interval, −4.50 to −2.98), indicating lower MCP-1/CCL2-related readouts in resveratrol-treated groups than in controls, with substantial heterogeneity (I2 = 78.9%). The negative association was driven mainly by rat and mouse datasets, whereas the piglet estimate was directionally opposite and the rabbit estimate came from a single dataset. Funnel-plot inspection suggested asymmetry, and dose meta-regression did not significantly explain between-study variation (slope = −0.17, p = 0.482). Leave-one-out and cumulative analyses indicated directional stability but did not resolve the underlying heterogeneity. Conclusions: These preclinical data indicate lower MCP-1/CCL2-related readouts after resveratrol treatment, but high heterogeneity, PubMed-only retrieval, and pharmacokinetic limitations limit direct clinical inference. Full article

Heart failure (HF) remains a major cause of morbidity and mortality despite substantial therapeutic progress, and important phenotype-specific treatment gaps persist. Protein S-nitrosylation (SNO) is a reversible cysteine-centered post-translational modification (PTM) whose reported associations with selected HF-relevant contexts, including vascular–endothelial dysfunction, mitochondrial–energetic remodeling, Ca²⁺-handling abnormalities, and selected receptor- or stress-related signaling observations, are supported to varying degrees. In this review, we evaluate reported mechanisms that may regulate cardiac SNO and define the evidentiary boundaries that constrain interpretation across HF-relevant settings. Available studies suggest that altered SNO homeostasis is associated with selected HF-related processes, but the strength of support varies substantially across targets, phenotypes, and disease contexts. Many mechanistic observations derive from animal models, cultured systems, donor-based perturbations, or non-HF settings. These should, therefore, be interpreted as hypothesis-generating rather than as established mechanisms in human HF. We accordingly distinguish findings supported by human HF tissue or HF-relevant in vivo evidence from more preliminary observations and highlight the need for human, site-resolved, and, where feasible, quantitatively grounded datasets. Future studies should prioritize stronger tissue anchoring, better integration of circulating and myocardial readouts, and closer alignment between mechanistic claims and the strength of the supporting evidence. Full article

Traditional rectal temperature measurement in pigs induces stress in animals, imposes a heavy labor burden on staff, and increases the risk of cross-infection. This study proposes a non-invasive deep learning approach to predict porcine rectal temperature by combining infrared thermal images of thermal windows with environmental parameters. A multimodal dataset is constructed by synchronously collecting thermal images, environmental parameters, and actual rectal temperatures. Mask Region-based Convolutional Neural Network (Mask R-CNN), You Only Look Once version 8 small (YOLOv8s), and YOLOv11s are employed to automatically detect or segment thermal window regions, from which the maximum temperature of each region is extracted. To enhance model generalization under varying environmental conditions, a two-stage hybrid regression framework is established. In this framework, a Convolutional Neural Network (CNN) extracts spatial features from thermal images, a fully connected network (FCNN) encodes regional surface temperatures and environmental parameters, and a Transformer module captures cross-modal dependencies to generate a preliminary prediction. Subsequently, a Random Forest (RF) regressor is applied for residual correction and final output optimization. Comparative experiments on single-region, dual-region, and triple-region combinations demonstrate that the “eye + vulva” dual-region scheme yields the optimal performance, with a mean absolute error (MAE) of 0.1796 °C and a coefficient of determination (R²) of 0.8212. The prediction error of this scheme is reduced by 42.3% compared with the best-performing unimodal model. The proposed method provides a fast, accurate, and stress-free solution for porcine body temperature monitoring, thereby supporting the development of intelligent health management in livestock farming. Full article

Myocardial fibrosis is a common pathological feature of multiple cardiovascular diseases, including heart failure, hypertension, and myocardial infarction, and is associated with poor prognosis. Despite extensive research, clinically validated molecular biomarkers for early diagnosis and reliable therapeutic targets for myocardial fibrosis remain limited. Post-translational modifications (PTMs), including phosphorylation, acetylation, ubiquitination, SUMOylation, and glycosylation, are critical regulators of fibrosis-related signaling pathways, yet a systematic bioinformatics-driven identification of PTM-related hub genes has not been performed. Three publicly available GEO datasets (GSE57345, GSE133054, GSE76314) comprising cardiac tissue from heart failure and control patients were integrated. Differentially expressed genes (DEGs) were identified using the limma package, then intersected with a curated PTM gene set derived from PhosphoSitePlus and UniProt databases. Weighted gene co-expression network analysis (WGCNA) identified fibrosis-associated modules, and protein–protein interaction (PPI) network analysis via STRING and CytoHubba pinpointed hub genes. Diagnostic performance was assessed by receiver operating characteristic (ROC) analysis across independent validation cohorts. Immune cell infiltration was estimated using CIBERSORT.Molecular docking with AutoDock Vina (version 1.2.3) was performed to evaluate binding affinity of FDA-approved cardiovascular drugs against identified hub protein targets. A total of 863 DEGs were identified in the training cohort (|log₂FC| > 1.0, adjusted p < 0.05), of which 138 overlapped with the PTM gene set. WGCNA revealed a turquoise module (r = 0.79, p < 0.001) most significantly correlated with fibrosis severity. PPI analysis identified five hub genes: SIRT3, SMAD3, NEDD4L, UBC9, and CAMK2D. ROC analysis demonstrated strong diagnostic performance (AUC range: 0.82–0.92) validated in independent cohorts. Hub genes showed significant correlations with M2 macrophage infiltration. Molecular docking identified spironolactone and finerenone as top-ranked ligands with binding energies of −8.7 and −8.4 kcal/mol against SMAD3 and SIRT3, respectively. This study, which is entirely in silico and based on publicly available transcriptomic datasets, systematically identifies five PTM-related hub genes as candidate diagnostic biomarkers and prioritised drug-repurposing targets in myocardial fibrosis. These findings are hypothesis-generating and require experimental validation (protein-level confirmation, cell- and animal-based functional assays, and biophysical binding studies) before any diagnostic or therapeutic claim can be made. Full article

To meet the requirements of accurate multi-class animal detection and model lightweighting in UAV-based grazing monitoring, this study presents DMN-YOLO, an efficient detector built upon YOLO11n. In particular, a lightweight downsampling module, DSDown, is introduced to alleviate the loss of detailed features of tiny targets during downsampling under complex grassland backgrounds, thereby improving the preservation of edge, texture, and local structural information. Meanwhile, a MACFPN multi-scale feature fusion structure is designed to handle large scale variations and feature confusion among multiple animal targets, enhancing cross-scale feature interaction and background suppression for better small-target representation. In addition, NWDR Loss combines CIoU geometric constraints, normalized Wasserstein distance, and an adaptive weighting strategy to improve overall stability and localization accuracy of small-target bounding box regression. Results indicate that DMN-YOLO attains 93.6% precision, 89.9% recall, and 95.8% mAP@0.5 on the UAV animal detection dataset. Compared with YOLO11n, it reduces the parameter count by 35.7% while lowering the model size by 29.3%. These results show that DMN-YOLO effectively reduces model complexity while maintaining strong detection performance, demonstrating good potential for practical field deployment. Full article

Cat behavior recognition in unconstrained videos is important for animal welfare monitoring and veterinary assessment, yet remains challenging because behavior cues are often carried by highly deformable and intermittently visible parts such as the head and tail. This study aims to improve clip-level cat behavior recognition under unstable part visibility in real-world videos. We propose PMTNet, a part-centric temporal network for cat behavior recognition under unstable part visibility. The framework first detects the cat body, head, and tail using a DEIM-based detector, then selects a detector according to video-domain continuity and stability, and finally models behavior from ROI appearance features and explicit geometric motion cues. The framework was developed and evaluated using a part-detection dataset of 4000 training images and 500 validation images, together with a cat behavior dataset of 1283 video clips across five categories. In the best-performing setting, PMTNet achieved 93.1% Top-1 Accuracy and 90.9% Macro-F1. Ablation studies further suggest that detector choice in the video domain, complementary part cues, and missing-aware fusion all contribute to the final recognition performance. On the present dataset, PMTNet also outperformed representative end-to-end video recognition baselines. These results support the use of part-centric temporal modeling for cat behavior recognition in unconstrained real-world videos. Full article

The non-invasive determination of live weight and body composition of ewes is an important element in ensuring precision livestock management and animal well-being. Traditional practices tend to be subjective, labor-intensive, or rely on expensive medical imaging such as Computed Tomography (CT). This paper proposes a new hybrid deep learning method to predict live weight and carcass traits in Coopworth ewes. The dataset of 1184 images taken from 156 ewes was analyzed and compared using a hybrid model (ResNet18 with Multi-Layer Perceptron through simple concatenation) and two more advanced models: Attention-Guided Feature Fusion Network (AGFF-Net) based on cross-modal attention and a Vision Transformer-based Hybrid Regressor (ViT-HR). Auxiliary tabular variables are the Body Condition Score (BCS) and size category. The Transformer architecture predicts (R² = 0.93) the live weight of ewes by dynamically ranking each visual patch and asking it to query the self-attention sequence. This technique treats the BCS as a distinct token in the self-attention sequence. Data partitioning at the animal level was stringent, thereby giving strong generalization. Findings indicate that the best advanced fusion systems are far better than baseline concatenation, with a high accuracy confirmed with gold standards obtained by CT. Grad-CAM visual explainability makes sure that models are able to localize biologically relevant anatomical locations successfully. The study closes the gap between complex deep learning models and real-world agriculture implementation to provide a correct, interpretable and scalable solution to real-time livestock measurements. Full article

Preclinical multimodal imaging is widely applied in small animal models for longitudinal studies of human diseases. Beyond murine systems, cost-effective and ethically sustainable models such as the chicken embryo and its chorioallantoic membrane are gaining increasing interest in accordance with the 3Rs principles. This study evaluated the feasibility of using both high-frequency ultrasound and magnetic resonance imaging for the non-invasive longitudinal monitoring of chicken embryo development in ovo. Fifty fertilized eggs were incubated under controlled conditions and examined up to embryonic day 14. High-frequency ultrasound (15–71 MHz) enabled real-time imaging and quantitative assessment of superficial structures, including cranial biometry and limb growth, while magnetic resonance imaging (7T) provided high-resolution three-dimensional visualization of internal organs and extraembryonic compartments. Together, these modalities allowed the progressive identification of key anatomical structures from ED5 onward, with HFUS enabling earlier linear measurements and MRI facilitating detailed anatomical and volumetric evaluation. The integration of these techniques allowed the generation of a developmental imaging timeline and quantitative reference dataset of normal embryogenesis. This multimodal approach represents a promising strategy for in vivo developmental studies, offering a robust baseline to characterize structural alterations induced by experimental conditions. Moreover, the use of the chicken embryo model provides significant ethical and economic advantages, supporting its application in preclinical research and imaging-based studies. Full article

Ketosis is a prevalent metabolic disorder in early-lactation dairy cows, significantly affecting animal health, milk production, and farm profitability. Developing accurate and non-invasive methods for early risk detection is therefore of critical importance. In this study, a hybrid optimization framework integrating an Improved Whale Migration Algorithm (IWMA) with a Light Gradient Boosting Machine (LightGBM) is proposed to predict ketosis risk based on the milk fat-to-protein ratio (F/P) using Dairy Herd Improvement (DHI) records. The proposed IWMA enhances optimization performance through cubic chaotic initialization, elite opposition-based learning, and a Cauchy–Gaussian hybrid mutation strategy, enabling improved global exploration and convergence stability. A dataset comprising 25,155 DHI records collected from multiple commercial dairy farms over seven months was used for model development and evaluation. Experimental results demonstrate that the IWMA–LightGBM model achieves a classification accuracy of 0.8997 and a mean squared error of 0.289, consistently outperforming six benchmark optimization methods. Feature analysis identifies Herd Within Index (WHI), Energy Corrected Milk (ECM), Days in Milk (DIM), Milk Urea Nitrogen, and Foremilk as key predictors associated with metabolic risk. Overall, the proposed approach provides a robust and effective non-invasive solution for early-stage metabolic risk screening at the herd level, offering practical value for precision dairy management. It should be noted that the model is intended for risk assessment rather than clinical diagnosis of ketosis. Full article

Thoracic impedance (TI) correlates with end-tidal carbon dioxide (ETCO₂) in large animals. This pilot study in Sprague-Dawley rats investigated whether TI can estimate ETCO₂ and dynamically guide compression depth. A dataset of TI and ETCO₂ measurements in rats was established to analyze the correlation between the two and construct a regression model. TI peak was strongly and positively correlated with ETCO₂ (r = 0.78, p < 0.001) and exhibited a progressive decay during prolonged compression. This pilot study demonstrated the feasibility of using TI-estimated ETCO₂ to guide compression depth in a rat model. The TI-guided strategy maintained ETCO₂ closer to the target value of 20 mmHg; however, no significant differences were observed between groups in ROSC rates, survival rates, blood gas parameters, or histopathological damage. Larger-scale studies are needed to evaluate clinical efficacy. Full article

Automatic classification of anime character painting styles is of great significance to the digital cultural industry and visual content production. Existing methods are prone to shortcut learning when handling complex color rendering and cannot fully decouple high-frequency line drafts from low-frequency colors. To solve this problem, this study proposes an improved deep learning classification method based on EfficientNetV2-B0. This method introduces random amplitude scaling (RAS) at the data input terminal. It realizes effective decoupling of colors and line-draft structures through random low-frequency amplitude perturbation, and suppresses the model’s excessive dependence on global color information from the source. Edge-guided coordinate attention (EG-CA) is integrated into the backbone network. It enhances the perception of line and contour features through edge weights and improves the model’s ability to capture fine-grained structural features. Adaptive scale feature aggregation (ASFA) is designed in the multi-scale feature fusion stage. It achieves efficient fusion of shallow textures and deep semantics through dynamic weighting, so as to enhance the model’s discriminative ability under complex painting styles. On a dataset containing 7887 images of four categories, the classification accuracy of the model reaches 95.81%. It significantly outperforms mainstream models such as MViTv2-T. Meanwhile, the number of parameters is only 7.84 M and the inference speed reaches 68.83 FPS. Ablation experiments show that the synergistic effect of the three modules improves the accuracy of the baseline model by 6.06%. It proves that the proposed method provides reliable technical support for the structured management and copyright traceability of anime images. Full article

Attributed graphs have recently emerged as a powerful tool for representing diverse data in numerous real-world sensors. Among various applications, unsupervised graph anomaly detection (UGAD) aims to identify abnormal data that significantly deviate from the majority of normal nodes without label annotations. Hence, UGAD can provide crucial assistance in enhancing the reliability of IoT, intelligent sensors and so on. Under the class-imbalanced reality caused by anomaly scarcity, the common paradigm of UGAD focuses on learning a model that primarily captures normal patterns. However, the traditional Graph Neural Network (GNN) paradigm suffers from local-aggregation limitations and over-smoothing, constraining their discrimination capacity. To address these issues, we introduce Graph Transformers (GTs) into UGAD task, termed as unsupervised attributed graph Anomaly detectioN wIth Masked grAph TransformErs (ANIMATE). Leveraging the global receptive field of Transformers, we can capture graph information that preserves the distinguishable characteristics of abnormalities from a global perspective. Furthermore, we employ masked auto-encoders to reconstruct node features and prompt our model to focus more on learning normal patterns. Additionally, we enhance the performance through a self-paced enhancement scheme specifically for UGAD tasks. Experiments conducted on various real-world benchmark datasets with organic anomalies validate the effectiveness of our proposed method compared to state-of-the-art competitors. Full article

Facial expression recognition (FER) is a well-established task in computer vision, yet its application to non-photorealistic domains, such as anime and manga, remains largely underexplored. The stylized, exaggerated, and often non-proportional facial features of illustrated characters present unique challenges for deep learning models trained predominantly on realistic imagery. In this work, we construct a balanced dataset of 3000 manga and anime face images spanning six emotion categories (Angry, Embarrassed, Happy, Manic–Euphoric, Sad, Scared) and conduct a systematic comparison of two major deep learning paradigms: Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs). Specifically, we evaluate ResNet-18, ResNet-50, ViT-B/16, and ViT-S/16 under four fine-tuning strategies: linear probing, partial fine-tuning, full fine-tuning, and progressive unfreezing, enabling a controlled comparison of both architectural families and transfer learning depth. Our results show that fine-tuning strategy significantly impacts performance: the best configuration (ViT-B/16 with progressive unfreezing) achieves 81.33% test accuracy (single run, seed 42), compared to 61.33% for the weakest linear probe baseline (ViT-S/16), a gap of 20.00 percentage points. To isolate architectural differences from strategy effects, we note that under full fine-tuning, the only strategy applied identically to all four models, ViT-S/16 (76.00%) outperforms ResNet-18 (74.44%) by 1.56 percentage points and ViT-B/16 (74.22%) by 1.78 percentage points, confirming a modest but consistent architectural advantage for Transformers once backbone adaptation is permitted. Vision Transformers benefit disproportionately from fine-tuning, and the relative ranking of architectures changes across fine-tuning regimes. Confusion matrix analysis reveals persistent cross-class confusion between visually similar emotions (e.g., Happy vs. Embarrassed), while the highly distinctive Manic–Euphoric category is consistently well recognized across all architectures. To the best of our knowledge, this is the first work to conduct a controlled multi-architecture, multi-strategy transfer learning benchmark specifically for FER in anime and manga, revealing findings that are not predictable from photographic FER literature and that carry direct practical implications for model selection in non-photorealistic visual recognition tasks. The anime and manga domain provides a uniquely controlled testbed for studying transfer learning under deliberate stylization, where the domain gap from realistic imagery is not an artifact of image degradation or environmental noise but a principled artistic choice with codified visual conventions; observing that fine-tuning depth dominates architectural choice in this domain suggests the same conclusion likely holds in other non-photorealistic transfer scenarios such as medical illustrations, architectural drawings, and synthetic training data. Full article

Background: Septic cardiomyopathy (SCM) is a fatal sepsis-induced dysfunction. While Pueraria (Pue) exhibits protective effects in sepsis, its regulatory role regarding programmed cell death (PCD) in SCM remains unclear. This study aimed to identify Pue’s PCD-related targets in SCM and construct a validated diagnostic model. Methods: We analyzed 14 PCD modalities across seven GEO transcriptomic datasets. A robust machine learning framework integrating 171 algorithm combinations built a diagnostic signature. The immune landscape was profiled using single-cell RNA sequencing and enrichment analyses. Experimental validation utilized SCM patient blood samples and heart tissues from an LPS-induced murine model. Results: Nine PCD patterns were significantly altered in SCM. Intersection analysis and machine learning identified five core Pue targets: STAT3, RIPK2, GM2A, ALOX5, and DPP4. A diagnostic model constructed with these genes achieved high AUCs across all datasets. Single-cell analysis revealed cell-type-specific expression within the myocardial immune landscape. Differential expression of these five genes was validated in both human and animal samples, correlating significantly with cardiac function indices. Conclusions: Our results demonstrate that Pueraria mitigates SCM and restores cardiac function by modulating the expression of core PCD-related targets. These targets are closely associated with the localized inflammatory response, providing potential therapeutic avenues for SCM. Full article