Hasil untuk "Dairying"

Gabriella Condello, Flavio S. Schenkel, Isis C. Hermisdorff et al.

ABSTRACT: Bovine spastic syndrome, known as Crampy, is a neuromuscular disorder in cattle. Affected cattle, 2 yr or older, suffer from involuntary muscle spasms in their hind limbs, leading to discomfort and reduced mobility. This often results in early culling from the herd, causing substantial financial loss for producers. Given the welfare implications and economic burden associated with Crampy, it is crucial to identify effective strategies to mitigate its occurrence. In this study, we assessed the feasibility of genetic selection to reduce Crampy by estimating variance components, evaluating the effect of incorporating genomic information, investigating Crampy's relationship with other economically important traits, and identifying genomic regions associated with Crampy in Canadian Holstein dairy cattle. A dataset comprising 54,826 animals, including 1,952 cases of Crampy, from 678 Canadian dairy herds, was provided by Lactanet Canada (Guelph, ON, Canada). Of these animals, 22,408 (including 408 with Crampy) were genotyped. Both threshold and linear models were used to estimate variance components, with observed scale h2 estimates ranging from 0.057 to 0.085. The inclusion of genomic data significantly increased the reliability of breeding values by 5% to 17%. Through a GWAS using GCTA software, a total of 41 significant SNPs were found to be significantly associated with Crampy. Functional analysis revealed 44 genes, among which we have highlighted the genes WNK2 (BTA8), DTNBP1 (BTA23), and ADK (BTA28), which have been associated with ion transport, muscle function, and neuron signaling, respectively. Enriched colocated QTL annotations linked to ketosis, muscle calcium content, and muscle zinc content were also identified, highlighting the role of metabolic processes and mineral homeostasis in muscle function. Breeding value correlations between Crampy and production, health, longevity, and type traits, and the selection indices were moderately low but favorable, indicating that current breeding strategies may indirectly select against Crampy. These findings highlight genomic selection as a viable strategy to mitigate Crampy in Canadian dairy herds, emphasizing the need for continued phenotyping for this disorder and optimization of breeding practices to improve animal welfare and sustainability.

Phoenix Yu, Tilo Burghardt, Andrew W Dowsey et al.

Holstein-Friesian detection and re-identification (Re-ID) methods capture individuals well when targets are spatially separate. However, existing approaches, including YOLO-based species detection, break down when cows group closely together. This is particularly prevalent for species which have outline-breaking coat patterns. To boost both effectiveness and transferability in this setting, we propose a new detect-segment-identify pipeline that leverages the Open-Vocabulary Weight-free Localisation and the Segment Anything models as pre-processing stages alongside Re-ID networks. To evaluate our approach, we publish a collection of nine days CCTV data filmed on a working dairy farm. Our methodology overcomes detection breakdown in dense animal groupings, resulting in a 98.93% accuracy. This significantly outperforms current oriented bounding box-driven, as well as SAM species detection baselines with accuracy improvements of 47.52% and 27.13%, respectively. We show that unsupervised contrastive learning can build on this to yield 94.82% Re-ID accuracy on our test data. Our work demonstrates that Re-ID in crowded scenarios is both practical as well as reliable in working farm settings with no manual intervention. Code and dataset are provided for reproducibility.

Dafne Zorzetto, Zizhao Xie, Julian Stamp et al.

Diet plays a crucial role in health, and understanding the causal effects of dietary patterns is essential for informing public health policy and personalized nutrition strategies. However, causal inference in nutritional epidemiology faces several challenges: (i) high-dimensional and correlated food/nutrient intake data induce massive treatment levels; (ii) nutritional studies are interested in latent dietary patterns rather than single food items; and (iii) the goal is to estimate heterogeneous causal effects of these dietary patterns on health outcomes. We address these challenges by introducing a sophisticated exposure mapping framework that reduces the high-dimensional treatment space via factor analysis and enables the identification of dietary patterns. We also extend the Bayesian Causal Forest to accommodate three ordered levels of dietary exposure, better capturing the complex structure of nutritional data and enabling estimation of heterogeneous causal effects. We evaluate the proposed method through extensive simulations and apply it to a multi-center epidemiological study of Hispanic/Latino adults residing in the US. Using high-dimensional dietary data, we identify six dietary patterns and estimate their causal link with two key health risk factors: body mass index and fasting insulin levels. Our findings suggest that higher consumption of plant lipid-antioxidant, plant-based, animal protein, and dairy product patterns is associated with reduced risk.

Bo Jin, Shichao Zhao, Jin Lyu et al.

The lactation performance of Saanen dairy goats, renowned for their high milk yield, is intrinsically linked to their body size, making accurate 3D body measurement essential for assessing milk production potential, yet existing reconstruction methods lack goat-specific authentic 3D data. To address this limitation, we establish the FemaleSaanenGoat dataset containing synchronized eight-view RGBD videos of 55 female Saanen goats (6-18 months). Using multi-view DynamicFusion, we fuse noisy, non-rigid point cloud sequences into high-fidelity 3D scans, overcoming challenges from irregular surfaces and rapid movement. Based on these scans, we develop SaanenGoat, a parametric 3D shape model specifically designed for female Saanen goats. This model features a refined template with 41 skeletal joints and enhanced udder representation, registered with our scan data. A comprehensive shape space constructed from 48 goats enables precise representation of diverse individual variations. With the help of SaanenGoat model, we get high-precision 3D reconstruction from single-view RGBD input, and achieve automated measurement of six critical body dimensions: body length, height, chest width, chest girth, hip width, and hip height. Experimental results demonstrate the superior accuracy of our method in both 3D reconstruction and body measurement, presenting a novel paradigm for large-scale 3D vision applications in precision livestock farming.

Fangjing Li, Zhihai Wang, Xinxin Ding et al.

Mounting posture is an important visual indicator of estrus in dairy cattle. However, achieving reliable mounting pose estimation in real-world environments remains challenging due to cluttered backgrounds and frequent inter-animal occlusion. We present FSMC-Pose, a top-down framework that integrates a lightweight frequency-spatial fusion backbone, CattleMountNet, and a multiscale self-calibration head, SC2Head. Specifically, we design two algorithmic components for CattleMountNet: the Spatial Frequency Enhancement Block (SFEBlock) and the Receptive Aggregation Block (RABlock). SFEBlock separates cattle from cluttered backgrounds, while RABlock captures multiscale contextual information. The Spatial-Channel Self-Calibration Head (SC2Head) attends to spatial and channel dependencies and introduces a self-calibration branch to mitigate structural misalignment under inter-animal overlap. We construct a mounting dataset, MOUNT-Cattle, covering 1176 mounting instances, which follows the COCO format and supports drop-in training across pose estimation models. Using a comprehensive dataset that combines MOUNT-Cattle with the public NWAFU-Cattle dataset, FSMC-Pose achieves higher accuracy than strong baselines, with markedly lower computational and parameter costs, while maintaining real-time inference on commodity GPUs. Extensive experiments and qualitative analyses show that FSMC-Pose effectively captures and estimates cattle mounting pose in complex and cluttered environments. Dataset and code are available at https://github.com/elianafang/FSMC-Pose.

A.L. Freihofer, R.M. Bruckmaier, J.J. Gross

ABSTRACT: In addition to the metabolic and health status of cows at parturition, intrauterine conditions, the calving process, and colostrum feeding may affect endocrine and metabolic pathways in the neonate. Forty-six clinically healthy cows without dystocia were enrolled, along with their calves. Blood samples were collected from cows (4 h postpartum [p.p.]) and calves (4, 12, and 24 h p.p.). Calves were fed colostrum from their dams immediately after blood sampling at 4 and 12 h p.p. Concentrations of glucose, insulin, nonesterified fatty acids (NEFA), IGF-1, and prolactin (PRL), as well as activities of aspartate-aminotransferase and gamma-glutamyltransferase were measured in the plasma of cows. In calves, we measured various endocrine and metabolic parameters related to protein, glucose, and lipid metabolism (e.g., NEFA, phospholipids [PL], total cholesterol [TC], low-density lipoprotein cholesterol [LDL-C], high-density lipoprotein cholesterol [HDL-C]), IgG, glucose, glucagon, and insulin. Pearson correlation coefficients among parameters measured in cows and calves were calculated. At 4 h p.p. (i.e., before colostrum feeding), maternal glucose was positively correlated with glucose (r = 0.29) and NEFA in calves (r = 0.25). Plasma NEFA in dams was negatively correlated with fat metabolism (PL: r = −0.31, HDL-C: r = −0.32) and plasma IgG (r = −0.28) in calves at 4 h p.p. Positive correlations were identified between the glucose of dams and calves (12 h p.p.: r = 0.26; 24 h p.p.: r = 0.45). Maternal NEFA was positively associated with calf lipid metabolism at 24 h p.p. (PL: r = 0.44, TC: r = 0.39, LDL-C: r = 0.37, HDL-C: r = 0.36). Primarily positive and significant correlations were detected between maternal PRL and lipid metabolism-related parameters in calves (NEFA at 12 h p.p.: r = 0.26, PL at 12 h p.p.: r = 0.31; PL at 24 h p.p.: r = 0.57, TC at 24 h p.p.: r = 0.62, LDL-C at 24 h p.p.: r = 0.48, and HDL-C at 24 h p.p.: r = 0.69). In conclusion, the metabolic status of neonates is partly associated with the metabolism of their mothers before the first feeding, whereas later associations between cow and calf are likely due to colostrum feeding.

Gebreyohans Gebru, Gebregergs Tesfamaryam, Dawit Gebremichael et al.

Calf morbidity and mortality pose significant economic challenges for smallholder dairy farms in Ethiopia, resulting in direct losses from calf deaths, replacement costs, treatment expenses, and reduced lifetime productivity. This study aimed to comprehensively investigate the magnitude and epidemiological characteristics of calf morbidity and mortality in Northern Ethiopia. A cross-sectional study with mixed approaches was carried out from December 2019 to September 2020. A total of 183 questionnaire survey, four focus group discussion (FGD), and 17 key informant interviews (KII) were included in the study. Furthermore, participatory epidemiological appraisals were incorporated to triangulate and strengthen survey evidences. Analysis of survey results revealed that 69.4% of the farmers have experienced calf morbidity, while 63.9% of them have encountered calf mortality. Similarly, results of proportional piling indicated that calf morbidity and mortality were estimated to occur in 75.5% and 55.9% of the farms, respectively. Moreover, all KIIs had encountered calf morbidity, while 88.2% of them had faced calf mortality. Ninety percent of KIIs, 66.2% of the participants of community-based epidemiology, and 27.87% of questionnaire survey respondents suggested that calf morbidity and mortality occur in less than one-week-aged calves. Regarding the potential risk factors, source of water, frequency of barn disinfection, breed types, health status of dams, using separate calf housing, amount of colostrum provided to calves, and cleaning frequency of barns had statistically significant association with the occurrence of calf morbidity and mortality (p<0.05). Additionally, results of participatory appraisal, FGDs and KIIs showed that calf diarrhea, nutritional disorder, pneumonia, and navel ill were the leading causes of calf morbidity and mortality. Furthermore, observation assessment showed that most dairy farms were surrounded by dense human settlements, livestock markets, and municipal slaughtering houses. Hence, the farms had critical space limitation (for animals to exercise) as well as poor drainage systems and hygienic practices. Our assessment also showed that lack of veterinary services, shortage of water supply, and poor artificial insemination services were the major challenges of dairying in the area. In conclusion, the present study revealed that calf morbidity and mortality were critical challenges for dairying in Northern Ethiopia. Furthermore, the study highlighted the epidemiological characteristics and potential risk factors associated with calf morbidity and mortality, awareness gaps in calf management, as well as key bottlenecks in dairy farming. These findings underscore the need for a comprehensive study, continuous capacity building initiatives, improved infrastructure, and services to mitigate calf losses.

Ana S. Cardoso, Sandra Martínez-Jarquín, Robert M. Hyde et al.

ABSTRACT: Lameness, defined as an impaired gait, impacts cow welfare and performance, compromising future health and production, and increasing culling risk. Untargeted milk lipidomics, together with the use of machine learning methods, have shown promise in identifying potential biomarkers for the early detection of lameness, before the development of visible clinical lameness. Prediction of early lameness would allow for the earlier implementation of management and treatment strategies, ultimately reducing the negative consequences. This study aimed to evaluate the predictive accuracy of differences in the milk metabolome and identify milk lipid biomarkers for early lameness detection in first-lactation dairy cows. Untargeted lipidomics and machine learning approaches were used to evaluate the differences in the milk metabolomic profiles in samples collected from heifers during the transition period (before lameness) and at the time of first lameness onset. A total of 56 milk samples from 32 cows (16 lame, 16 control) were analyzed by liquid chromatography–high-resolution mass spectrometry after calving (before lameness) and at lameness onset. Elastic net regression achieved 83% accuracy in predicting lameness from samples collected after calving and 100% accuracy at the time of lameness. A total of 10 mass ions selected by different statistical methods showed potential to be considered predictors of lameness. Pathway analysis revealed significant dysregulation of retinol metabolism after calving in cows that go on to develop lameness in that lactation. This study demonstrated potential for using milk lipidomics for early lameness detection. This, in turn, provides insights into lameness pathogenesis, furthering our understanding of lameness, with the ultimate goal of developing interventions to improve dairy cow welfare and farm productivity.

C. Benchaar, P. Denis, P.Y. Chouinard

ABSTRACT: The objective of this study was to investigate the effects of dietary inclusion of different sources of unsaturated vegetable oil on ruminal fermentation characteristics, nutrient digestion, enteric CH4 emissions, nitrogen partitioning, milk production, and milk fatty acid composition. Eight multiparous ruminally cannulated Holstein cows were used in a duplicated 4 × 4 Latin square design (28-d periods) balanced for residual effects. Treatments consisted of a TMR based on alfalfa silage and corn silage (forage/concentrate ratio 52:48; DM basis) either not supplemented (CTL), or supplemented (4% of DM) with either sunflower (SUN, 67.5% cis-9 18:1), safflower (SAF, 74.6% cis-9, cis-12 18:2), or linseed (LSO, 54.9% cis-9,cis-12,cis-15 18:3) oil. The inclusion of the vegetable oil was performed primarily at the expense of another energy source (i.e., corn grain). Feeding the oil-supplemented diets decreased DMI (−1.2 kg/d, on average) and increased apparent total-tract digestibility of CP, compared with the CTL. However, no changes were noted in digestibility of DM, OM, NDF, gross energy, or the effective ruminal degradability of DM (i.e., in situ technique). Milk yield decreased only when feeding SUN (−3.3 kg/d) or SAF (−4.1 kg/d), compared with the CTL. Milk fat yield decreased when using SAF (−0.44 kg/d, −38%), SUN (−0.32 kg/d, −28%), and LSO (−0.21 kg/d, −18%), compared with the CTL. Milk protein yield was reduced with SUN (−9%) and SAF (−7%) compared with the CTL but was unaffected by LSO. The content of MUN decreased with oil supplementation by 28%, 20%, and 10% for LSO, SAF, and SUN, respectively. In contrast, the efficiency of dietary N use for milk N secretion was not affected when cows were fed the oil-supplemented diets. Relative to N intake, fecal N decreased, whereas urinary N increased with oil-supplemented diets. Oil supplementation did not alter enteric daily CH4 production (SF6 tracer technique) or yield. In the present study, the oils differing in unsaturation level did not change overall ruminal fermentation characteristics, OM digestibility, or CH4 production. Feeding the oil-based diets had no effect on N balance but shifted N excretion from feces to urine. Cows consuming the oil-supplemented diets exhibited reduced milk fat yield, depending on the unsaturation level, with a more pronounced decrease with SAF (i.e., a source of cis-9,cis-12 18:2), followed by SUN (i.e., a source of cis-9 18:1) and LSO (i.e., a source of cis-9,cis-12,cis-15 18:3).

Mohammed Temraz, Mark T Keane

In recent years, humanity has begun to experience the catastrophic effects of climate change as economic sectors (such as agriculture) struggle with unpredictable and extreme weather events. Artificial Intelligence (AI) should help us handle these climate challenges but its most promising solutions are not good at dealing with climate-disrupted data; specifically, machine learning methods that work from historical data-distributions, are not good at handling out-of-distribution, outlier events. In this paper, we propose a novel data augmentation method, that treats the predictive problems around climate change as being, in part, due to class-imbalance issues; that is, prediction from historical datasets is difficult because, by definition, they lack sufficient minority-class instances of "climate outlier events". This novel data augmentation method -- called Counterfactual-Based SMOTE (CFA-SMOTE) -- combines an instance-based counterfactual method from Explainable AI (XAI) with the well-known class-imbalance method, SMOTE. CFA-SMOTE creates synthetic data-points representing outlier, climate-events that augment the dataset to improve predictive performance. We report comparative experiments using this CFA-SMOTE method, comparing it to benchmark counterfactual and class-imbalance methods under different conditions (i.e., class-imbalance ratios). The focal climate-change domain used relies on predicting grass growth on Irish dairy farms, during Europe-wide drought and forage crisis of 2018.

You Chang, Jose L. Gonzales, Erik Rattenborg et al.

Since early 2024, highly pathogenic avian influenza virus (HPAIV) H5N1 of clade 2.3.4.4b has spilled over from wild birds to dairy cattle in the United States (U.S.), spreading to more than 1000 herds and threatening both animal and public health. Denmark's location along major migratory flyways and the lack of active HPAIV surveillance in cattle underscore the need to assess potential spillover risk from wild birds to cattle to strengthen preparedness. A quantitative spillover risk assessment model was developed to integrate data from Bird Flu Radar, eBird, and cattle density to estimate the weekly probability of HPAIV introduction from wild birds to cattle. The model was calibrated using observed U.S. spillover data and extrapolated to Denmark under the assumption of a comparable transmission rate parameter. Under the frequency-dependent model, the expected HPAIV introductions to Danish cattle via wild birds remain below 0.35 cases per week, with the highest temporal risk from December to March. High-risk areas were concentrated along the Danish coastline and near the German border. In contrast, applying a density-dependent model shifted the spatial risk toward regions with higher cattle densities, while the high-risk temporal periods remained unchanged. Overall, the results indicate a spatially and temporally variable risk of HPAIV spillover from wild birds to cattle in Denmark. The model establishes a data-driven framework to strengthen early warning and guide targeted surveillance efforts in high-risk regions.

Justin G. Bendall, Abraham S. Chawanji, Bertram Y. Fong et al.

ABSTRACT: Whole milk powder (WMP) manufactured in New Zealand in 1907 was sent to the Antarctic continent with the Shackleton-led British Antarctic Expedition from 1907 to 1909. This powder was stored at ambient conditions at Shackleton's Hut at Cape Royds, Antarctica, for over 100 yr before a sample was collected on behalf of Fonterra by the Antarctic Heritage Trust. Having spent most of its existence both dried and in frozen storage, any deleterious reactions within the WMP would have been markedly retarded. The composition and some properties of the roller-dried Shackleton's WMP are reported along with those of 2 modern spray-dried New Zealand WMP. The Shackleton powder was less white and more yellow than the modern WMP and was composed of flakes rather than agglomerated particles, consistent with that expected of a roller-dried powder. Headspace analysis showed lipolytic and oxidative volatile compounds were present in the Shackleton WMP, indicting some deterioration of the milk either before powder manufacture or on storage of the finished product. On a moisture-free basis, the Shackleton WMP had higher protein, higher fat (with a markedly higher free fat level), higher ash, and a lower lactose level than the modern WMP. The lysine level was lower in the Shackleton WMP compared with the spray-dried powders, whereas the fatty acid composition was relatively similar. The sodium level was markedly higher in the Shackleton WMP compared with the spray-dried powder, which is probably due to the addition of an alkaline sodium salt to adjust the pH of the milk before roller drying. Lead, iron, and tin levels were markedly higher in the Shackleton WMP compared with the spray-dried powders, possibly due to the equipment used in powder manufacture and the tin-plated cases used for storage. The proteins in the Shackleton WMP were more lactosylated than in the spray-dried powders. The Shackleton WMP had a higher ratio of κ-casein A to B variants and a higher ratio of β-lactoglobulin B to A variants than the spray-dried powders, whereas the αS1-casein, β-casein, αS2-casein, and α-lactalbumin protein variants were similar in all powders. The total phospholipid content was markedly lower in the Shackleton WMP than the spray-dried powders, primarily due to a lower phosphatidylethanolamine concentration. The molecular species distributions within the phospholipid classes were generally similar in the 3 powders. Claims are sometimes encountered that the milk of today is different from that consumed by previous generations. However, this comparative study has shown that the Shackleton WMP was generally similar to modern WMP. Although differences in some components and properties were observed, these were attributable to the manufacturing equipment and processes used in the pioneering years of WMP manufacture.

Babak Javadi, Zeinab Salimzadeh, Amir Hossein Akbari et al.

Considering the evolution of the food industry and its challenges, like high perishability, managing the food industry supply chain is a key focus for researchers and decision-makers. Uncertainty in decision-making has gained importance, particularly in the yogurt industry, known for its complexity. This study addresses production and distribution planning, scheduling, and routing in the yogurt supply chain. The problem is characterized by multiple products, a single plant, multiple distribution centers, multiple periods, and various transportation methods. A mixed-integer non-linear programming (MINLP) model is used to minimize total costs, including production, setup, overtime, unmet demand, and transportation. Additionally, a robust fuzzy programming approach is applied under uncertainty, with linearization procedures proposed to convert it into a linearized mixed-integer programming formulation. The problem is tested with two data types: a sample problem in three sizes (small, medium, and large) and real data from Kalle Dairy Company, Iran. A Genetic Algorithm (GA) is developed to solve the problem, with necessary modifications made for its application. The GA's performance is compared to an exact algorithm (Branch & Cut), showing that the company's production policy adapts daily to meet demand precisely. The shift to smaller batch production and longer shelf life allows better stock allocation and avoids shortages in uncertain conditions. The company's policies adapt to severe fluctuations in the business environment, though this requires high costs, such as inventory maintenance.

Neris Özen, Wenjuan Mu, Esther D. van Asselt et al.

The number of scientific articles published in the domain of food safety has consistently been increasing over the last few decades. It has therefore become unfeasible for food safety experts to read all relevant literature related to food safety and the occurrence of hazards in the food chain. However, it is important that food safety experts are aware of the newest findings and can access this information in an easy and concise way. In this study, an approach is presented to automate the extraction of chemical hazards from the scientific literature through large language models. The large language model was used out-of-the-box and applied on scientific abstracts; no extra training of the models or a large computing cluster was required. Three different styles of prompting the model were tested to assess which was the most optimal for the task at hand. The prompts were optimized with two validation foods (leafy greens and shellfish) and the final performance of the best prompt was evaluated using three test foods (dairy, maize and salmon). The specific wording of the prompt was found to have a considerable effect on the results. A prompt breaking the task down into smaller steps performed best overall. This prompt reached an average accuracy of 93% and contained many chemical contaminants already included in food monitoring programs, validating the successful retrieval of relevant hazards for the food safety domain. The results showcase how valuable large language models can be for the task of automatic information extraction from the scientific literature.

Azadeh Hassanpour, Johannes Geibel, Henner Simianer et al.

The effective planning and allocation of resources in modern breeding programs is a complex task. Breeding program design and operational management have a major impact on the success of a breeding program and changing parameters such as the number of selected/phenotyped/genotyped individuals will impact genetic gain, genetic diversity, and costs. As a result, careful assessment and balancing of design parameters is crucial, considering the trade-offs between different breeding goals and associated costs. In a previous study, we optimized the resource allocation strategy in a dairy cattle breeding scheme via the combination of stochastic simulations and kernel regression, aiming to maximize a target function containing genetic gain and the inbreeding rate under a given budget. However, the high number of simulations required when using the proposed kernel regression method to optimize a breeding program with many parameters weakens the effectiveness of such a method. In this work, we are proposing an optimization framework that builds on the concepts of kernel regression but additionally makes use of an evolutionary algorithm to allow for a more effective and general optimization. The key idea is to consider a set of potential parameterizations of the breeding program, evaluate their performance based on stochastic simulations, and use these outputs to derive new parametrization to test in an iterative procedure. The evolutionary algorithm was implemented in a Snakemake pipeline to allow for efficient scaling on large distributed computing platforms. The algorithm achieved convergence to the same optimum with a massively reduced number of simulations. Thereby, the incorporation of class variables and accounting for a higher number of parameters in the optimization pipeline leads to substantially reduced computing time and better scaling for the desired optimization of a breeding program.

Phoenix Yu, Tilo Burghardt, Andrew W Dowsey et al.

We present MultiCamCows2024, a farm-scale image dataset filmed across multiple cameras for the biometric identification of individual Holstein-Friesian cattle exploiting their unique black and white coat-patterns. Captured by three ceiling-mounted visual sensors covering adjacent barn areas over seven days on a working dairy farm, the dataset comprises 101,329 images of 90 cows, plus underlying original CCTV footage. The dataset is provided with full computer vision recognition baselines, that is both a supervised and self-supervised learning framework for individual cow identification trained on cattle tracklets. We report a performance above 96% single image identification accuracy from the dataset and demonstrate that combining data from multiple cameras during learning enhances self-supervised identification. We show that our framework enables automatic cattle identification, barring only the simple human verification of tracklet integrity during data collection. Crucially, our study highlights that multi-camera, supervised and self-supervised components in tandem not only deliver highly accurate individual cow identification, but also achieve this efficiently with no labelling of cattle identities by humans. We argue that this improvement in efficacy has practical implications for livestock management, behaviour analysis, and agricultural monitoring. For reproducibility and practical ease of use, we publish all key software and code including re-identification components and the species detector with this paper, available at https://tinyurl.com/MultiCamCows2024.

Nadia Innocente, Niccolò Renoldi, Erica Moret et al.

ABSTRACT: The possible contribution of brine-derived microflora to the sensory attributes of cheese is still a rather unexplored field. In this study, 365 bacteria and 105 yeast strains isolated from 11 cheese brines were qualitatively tested for proteolytic and lipolytic activities, and positive strains were identified by sequencing. Among bacteria, Staphylococcus equorum was the most frequent, followed by Macrococcus caseolyticus and Corynebacterium flavescens. As for yeasts, Debaryomyces hansenii, Clavispora lusitaniae, and Torulaspora delbrueckii were most frequently identified. A total of 38% of bacteria and 59% of yeasts showed at least 1 of the metabolic activities tested, with lipolytic activity being the most widespread (81% of bacteria and 95% of yeasts). Subsequently 15 strains of bacteria and 10 yeasts were inoculated in a curd-based medium and assessed via headspace-solid phase microextraction coupled with gas chromatography-mass spectrometry to determine their volatilome. After a 30-d incubation at 12°C, most strains showed a viability increase of about 2 log cfu/mL, suggesting good adaptability to the cheese environment. A total of 26 compounds were detected in the headspace, carbonyl compounds and alcohols being the major contributors to the volatile profile of the curd-based medium. Multivariate analysis was carried out to elucidate the overall differences in volatiles produced by selected strains. Principal component analysis and hierarchical clustering analysis demonstrated that the brine-related microorganisms were separated into 3 different groups, suggesting their different abilities to produce volatile compounds. Some of the selected strains have been shown to have interesting aromatic potential and to possibly contribute to the sensory properties of cheese.

S.G. Umaña Sedó, C.B. Winder, D.L. Renaud

ABSTRACT: Calf mortality can be used as an indicator of animal health and welfare on dairy farms. However, several challenges surround the estimation and reporting of this metric, specifically: (1) lack of records or reliable data, (2) methods of data collection, and (3) inconsistencies in calculation and definitions used. Therefore, despite its importance, the lack of consensus on a definition of calf mortality makes it difficult to compare mortality rates between dairy farms or studies. Monitoring factors associated with calf mortality is vital to create preventative strategies. Although common strategies have been set about how to raise dairy calves and manage dairy calves, discrepancies among studies evaluating factors associated with calf mortality still exist. This review summarizes research on the evaluation of calf mortality and associated risk factors, specifically, the lack of reliable data and standardization of the definition of calf mortality. In addition, current strategies to monitor and prevent calf mortality will be presented in this review.