Ahmed G et al. · Jul 1, 2026
Precision livestock farming (PLF) leverages activity sensors to monitor behaviours like grazing, resting and walking, yet class imbalance in datasets often leads to underrepresentation of minority behaviours such as 'escaping' and 'being mounted.' This study proposes a novel framework combining long short-term memory (LSTM) networks with the synthetic minority oversampling technique (SMOTE) to address this challenge. Unlike existing methods that use complex SMOTE variants such as DeepSMOTE or latent space augmentations, which add computational complexity and overhead, our approach integrates simple SMOTE with non-overlapping windowed segmentation, preserving sequential patterns during synthetic data generation while augmenting minority classes. The LSTM architecture captures temporal dependencies in the balanced dataset, enabling robust behaviour recognition. Evaluated on a composite accelerometer dataset derived from three distinct cows, the framework generalises across breeds, overcoming limitations of breed-specific models. It achieves state-of-the-art performance with 97.24% accuracy, 97.56% precision, 97.24% recall and a 97.29% F1-score, significantly improving detection of rare behaviours without compromising majority class precision. By unifying data from multiple cows, the model ensures robustness to behavioural variability, enhancing scalability for diverse farming environments. The simplicity of using basic SMOTE reduces computational overhead, making the solution practical for real-world deployment. This work bridges classical data balancing techniques with modern deep learning, offering a resource-efficient blueprint for handling imbalanced time-series data in agricultural AI. The results advance precision livestock farming by improving the reliability of automated behaviour monitoring, directly contributing to enhanced animal welfare and farm productivity through accessible, breed-agnostic AI tools.
Veterinary
Ko NT et al. · Jul 1, 2026
In humans, the ligamentum mucosum (LM) is described as a ligamentous structure originating from the femoral intercondylar notch and inserting into the infrapatellar fat pad. Proposed clinical implications include knee stabilization, contribution to post-operative revascularization of adjacent structures, and causation of anterior knee pain if inflamed. Published reports of the LM are rare in dogs and, to our knowledge, none exist for cats. Because common veterinary gross anatomy texts omit the LM from their descriptive anatomy of the stifle, students sometimes mistake the LM for the cranial cruciate ligament. Therefore, a description of the LM in the feline and canine stifle would serve as an important veterinary anatomy learning resource. The aim of this project was to characterize the LM in the cat and dog. Stifles were dissected from 62 cat hindlimbs (n = 24 preserved, n = 38 fresh) and 47 dog hindlimbs (n = 9 preserved, n = 38 fresh). The presence or absence of the LM was determined and described. Representative samples were processed for haematoxylin and eosin staining. Grossly, the LM was found bilaterally in 95.2% of cats and in 83.0% of canine limbs, appearing as an elastic, friable band of white-to-pink tissue tethering the infrapatellar fat pad to the femoral intercondylar notch. Histological samples revealed collagen fibrils, vascular structures and neural tissue. These data provide evidence of the LM in the cat and dog, and bolster currently available anatomic educational resources. The presence of the LM in the canine and feline stifle merits further investigation into its function in health and disease states.
Veterinary
Jafarbeglou M. · Jul 1, 2026
Intranasal (IN) drug delivery has increasingly considered as an easy, practical and non-invasive alternative to parenteral administration in veterinary medicine, offering rapid systemic and potential nose-to-brain effects. The first part of this review systematically collected and synthesized published evidence on IN administration across animal species, while the second part critically analysed the anatomical, pharmacological and technical factors that determine its success and limitations. Part I consisted a total of 110 eligible studies published between 1991 and 2025, encompassing dogs, cats, rabbits, pigs, ruminants, birds and reptiles. IN delivery has been investigated for a range of purposes and produced clinically meaningful sedation, analgesia and drug reversal, often comparable to intramuscular administration but generally characterized by slower onset and greater variability among species. Despite encouraging and favourable results, IN delivery was not without limitations. Its effectiveness can be strongly influenced by species-specific nasal anatomy and physiology, formulation characteristics and dosing volume. Defensive reactions, poor tolerability, sneezing, nasopharyngeal irritations, hypersalivation or swallowing of the drug are frequently reported. Future progress requires species-specific case selection guidelines and dosing standards, pharmacokinetic validation and developing safe concentrated formulations. Transparent reporting and balanced assessment of both benefits and drawbacks are essential to ensure the safe, effective and ethically responsible integration of IN administration into veterinary anaesthesia and critical care practice.
Veterinary
Raquib R et al. · Jul 1, 2026
Background Brucellosis remains an important zoonotic disease in Bangladesh, yet evidence on its prevalence across human and animal hosts is fragmented and heterogeneous. Objectives This study aims to synthesize evidence on Brucella detection across human and animal hosts in Bangladesh and to evaluate associated study-level factors and temporal patterns. Methods The data were extracted from eligible studies published between 2000 and 2024 following a systematic review framework. Pooled prevalence was estimated using a random-effects meta-analysis. Heterogeneity was assessed using Cochran's Q and I 2 statistics. Subgroup meta-analyses and univariable meta-regression were conducted to examine variation by host species, diagnostic method, and study-level factors. Forecasting was performed using ARIMA models on annual study-reported Brucella detection, with model performance evaluated using AIC, MAE, and RMSE. Results The pooled Brucella detection across hosts was 3.75% (95% CI: 3.07-4.56), with substantial heterogeneity (I 2 = 89.5%). Detection was the highest in dogs (7.49%) and the lowest in horses (1.78%)). Among evaluated variables, abortion history was the only study-level factorsignificantly associated with higher Brucella positivity, with aborted animals showing approximately 10 times higher odds of detection compared to non-aborted animals (OR = 10.09, 95% CI: 4.64-21.93, p Conclusion Brucella exposure and detection persist across multiple host species in Bangladesh, with variation largely driven by methodological and host-related differences. These findings underscore the need for standardized diagnostic and reporting approaches and support integrated One Health surveillance to improve interpretation and control strategies in Bangladesh.
Veterinary