Optical flow patterns in broiler chicken flocks as automated measures of behaviour and gait

Equivalence Between Optical Flow, the Unrest Index, and Walking Distance to Estimate the Welfare of Broiler Chickens

Modern poultry production demands scalable and non-invasive methods to monitor animal welfare, particularly as broiler strains are increasingly bred for rapid growth, often at the expense of mobility and health. This study evaluates two advanced computer vision techniques-Optical Flow and the Unrest Index-to assess movement patterns in broiler chickens. Three commercial broiler strains (Hybro®, Cobb®, and Ross®) were housed in controlled environments and continuously monitored using ceiling-mounted video systems. Chicken movements were detected and tracked using a YOLO model, with centroid data informing both the Unrest Index and distance walked metrics. Optical Flow velocity metrics (mean, variance, skewness, and kurtosis) were extracted using the Farnebäck algorithm. Pearson correlation analyses revealed strong associations between Optical Flow variables and traditional movement indicators, with average velocity showing the strongest correlation to walked distance and the Unrest Index. Among the evaluated strains, Cobb® demonstrated the strongest correlation between Optical Flow variance and the Unrest Index, indicating a distinct movement profile. The equipment's movement and the camera's slight instability had a minimal effect on the Optical Flow measurement. Still, its strong correlation with the Unrest Index and walking distance accredits it as an effective method for high-resolution behavioral monitoring. This study supports the integration of Optical Flow and Unrest Index technologies into precision livestock systems, offering a foundation for predictive welfare management at scale.

Genetics of gait score in broilers: Genetic parameters of gait score in purebred broiler lines

Leg health assessment is a vital component of poultry breeding goals; it includes a range of skeletal disorders and contact dermatitis traits as well as a general assessment of a bird's walking ability in the form of gait score (GS). If trained scorers are used, GS correlates well with more objective walking assessment methods and allows for high throughput data collection from large numbers of individuals. The paper describes for the first time GS heritabilities for broilers. The Aviagen GS system (AGS) consists of an 8-class scale of 10-45 in increments of 5, whereby 10 is the best score and 45 the worst. It has sufficient variation between categories to enable a clear distinction of GS classes for selection purposes. The results of GS at 4-5 weeks of age show GS heritabilities between 0.14 and 0.24 which is higher than turkey (0.08-0.13) or duck (0.06-0.12) heritabilities. Genetic correlations with body weight and breast percentage were unfavorable but moderate. Correlations among leg health traits (AGS, long bone deformities, crooked toes, hockburn and footpad dermatitis) were generally low (-0.16-0.10) for most traits and lines. While genetic correlations between AGS and production traits were unfavorable, both AGS and other leg health and production traits can be improved simultaneously in a multi-trait selection strategy combining production and welfare characteristics. AGS has been included in the Aviagen breeding goal since 2011.

Modern Technologies for Improving Broiler Production and Welfare: A Review

The increasing level of poultry meat production, the lack of human workforce, and the rapid development of information technology have led to the application of precision livestock farming (PLF) systems in the poultry sector, as in other livestock sectors. This review aimed to gather information on the function, applicability, advantages, and limitations of a wide range of precision technologies applicable to broiler production to help farms and researchers in choosing the right methods in practice or creating the basis for further development. Studies as the basis of this review were extracted from more than a hundred peer-reviewed articles including mainly publications of recognised journals and conference proceedings. The results showed that most precision tools are currently undergoing testing and focus on some parameters or a group of parameters which are closely related. Information on the prevalence of PLF systems on broiler farms is not available in the literature. The accessible technologies of different purposes should be combined and connected to enable communication with each other and create a complex, reliable, and precise background for farming. This also facilitates management decisions and the treatment of so-called "Big Data".

Top-view characterization of broiler walking ability and leg health using computer vision

Impaired walking ability and leg health are commonly seen in broilers and can negatively impact their welfare. Commonly, walking ability and leg health are assessed manually, but this is time consuming and can be subjective. Automated approaches for scoring walking ability and leg health at the individual level could therefore have great added value. Here, we studied whether automatically extracted top-view walking features of broilers can be used as a proxy for walking ability and leg health. Top-view videos were collected of unmarked broilers walking through a corridor that was placed inside the home pen. From these videos, four top-view features were derived: 1) lateral body oscillation, calculated as deviations from the movement trajectory of the bird, 2) step count, 3) completion time, and 4) length-width ratio of the virtual bounding box encapsulating the bird while walking as an indicator of wing support. We assessed the relationship of these computer vision-based features with manual gait and leg health scores, including hock burn (HB) and footpad dermatitis (FPD). We observed that birds with worse gait scores (GS) had longer completion times, higher step counts and a trend for higher lateral body oscillation levels in the walkway setup. Unsupervised clustering using the K-means algorithm with these walking features showed potential to distinguish birds with GS3+, although differentiating between GS1 and GS2 proved more challenging. We concluded that the length-width ratio of the bounding box during walking was not a suitable proxy for poor gait. We found no relationship between top-view walking features and mild cases of HB and FPD in broilers. Overall, the results of this study indicate that top-view video recordings can provide insight into birds' walking ability, using features related to movement speed, step count and lateral body oscillation, making automated scoring more feasible on a larger scale in practice. However, these top view features provide little information about mild HB and FPD.

Overview of the various methods used to assess walking ability in broiler chickens

BACKGROUND

Walking ability is a highly important animal-based indicator of welfare in broilers that warrants monitoring across commercial production.

METHODS

A review of the technical and scientific literature on existing methods for assessing walking ability in broilers was conducted, and the reliability, validity and feasibility of the different methods were assessed when relevant information was available.

RESULTS

Visual gait scoring has high validity, minimal equipment needs and is relatively quick to perform; however, this method is subjective and requires extensive training. Due to this, more objective tests, such as the latency to lie test, have been developed, although more validation of these tests is required. Other identified methods, such as obstacle courses, the rotarod test and kinetic and kinematic approaches, require considerable amounts of equipment, making them impractical for use during on-farm inspections. Automated methods that rely on movement or activity data have potential for future development but currently lack the resolution of traditional gait scoring and require equipment such as cameras or accelerometers.

CONCLUSIONS

Objective, seemingly valid, repeatable and feasible options exist for the assessment of walking ability that would facilitate the monitoring of commercial broiler welfare; however, more validation of these methods is required, especially in commercial farm settings.

Comparative Analysis of Broiler Housing Systems: Implications for Production and Wellbeing

This study compares the effects of modern colony cage systems and traditional floor systems on the production and welfare of broiler chickens. Through two trials spanning 35 days each, we evaluated various physiological parameters, including growth performance, bone health, stress responses, and meat quality. Colony cages demonstrated superior thermal regulation and growth performance compared to traditional floor systems, but also exhibited higher frequencies of leg deformity and reduced standing ability. Conversely, the broilers in traditional floor systems experienced heat stress-related challenges, impacting the meat quality. Our findings underscore the need to balance productivity with animal welfare in broiler farming practices. By understanding the distinct impacts of different housing systems, we can work towards improving broiler rearing methods to ensure optimal welfare and production outcomes.

Active walking in broiler chickens: a flagship for good welfare, a goal for smart farming and a practical starting point for automated welfare recognition

Automated assessment of broiler chicken welfare poses particular problems due to the large numbers of birds involved and the variety of different welfare measures that have been proposed. Active (sustained, defect-free) walking is both a universally agreed measure of bird health and a behavior that can be recognized by existing technology. This makes active walking an ideal starting point for automated assessment of chicken welfare at both individual and flock level.

Regulation of Meat Duck Activeness through Photoperiod Based on Deep Learning

The regulation of duck physiology and behavior through the photoperiod holds significant importance for enhancing poultry farming efficiency. To clarify the impact of the photoperiod on group-raised duck activeness and quantify duck activeness, this study proposes a method that employs a multi-object tracking model to calculate group-raised duck activeness. Then, duck farming experiments were designed with varying photoperiods as gradients to assess this impact. The constructed multi-object tracking model for group-raised ducks was based on YOLOv8. The C2f-Faster-EMA module, which combines C2f-Faster with the EMA attention mechanism, was used to improve the object recognition performance of YOLOv8. Furthermore, an analysis of the tracking performance of Bot-SORT, ByteTrack, and DeepSORT algorithms on small-sized duck targets was conducted. Building upon this foundation, the duck instances in the images were segmented to calculate the distance traveled by individual ducks, while the centroid of the duck mask was used in place of the mask regression box's center point. The single-frame average displacement of group-raised ducks was utilized as an intuitive indicator of their activeness. Farming experiments were conducted with varying photoperiods (24L:0D, 16L:8D, and 12L:12D), and the constructed model was used to calculate the activeness of group-raised ducks. The results demonstrated that the YOLOv8x-C2f-Faster-EMA model achieved an object recognition accuracy (mAP@50-95) of 97.9%. The improved YOLOv8 + Bot-SORT model achieved a multi-object tracking accuracy of 85.1%. When the photoperiod was set to 12L:12D, duck activeness was slightly lower than that of the commercial farming's 24L:0D lighting scheme, but duck performance was better. The methods and conclusions presented in this study can provide theoretical support for the welfare assessment of meat duck farming and photoperiod regulation strategies in farming.

Automated pose estimation reveals walking characteristics associated with lameness in broilers

Walking ability of broilers can be improved by selective breeding, but large-scale phenotypic records are required. Currently, gait of individual broilers is scored by trained experts, however, precision phenotyping tools could offer a more objective and high-throughput alternative. We studied whether specific walking characteristics determined through pose estimation are linked to gait in broilers. We filmed male broilers from behind, walking through a 3 m × 0.4 m (length × width) corridor one by one, at 3 time points during their lifetime (at 14, 21, and 33 d of age). We used a deep learning model, developed in DeepLabCut, to detect and track 8 keypoints (head, neck, left and right knees, hocks, and feet) of broilers in the recorded videos. Using the keypoints of the legs, 6 pose features were quantified during the double support phase of walking, and 1 pose feature was quantified during steps, at maximum leg lift. Gait was scored on a scale from 0 to 5 by 4 experts, using the videos recorded on d 33, and the broilers were further classified as having either good gait (mean gait score ≤2) or suboptimal gait (mean gait score >2). The relationship of pose features on d 33 with gait was analyzed using the data of 84 broilers (good gait: 57.1%, suboptimal gait: 42.9%). Birds with suboptimal gait had sharper hock joint lateral angles and lower hock-feet distance ratios during double support on d 33, on average. During steps, relative step height was lower in birds with suboptimal gait. Step height and hock-feet distance ratio showed the largest mean deviations in broilers with suboptimal gait compared to those with good gait. We demonstrate that pose estimation can be used to assess walking characteristics during a large part of the productive life of broilers, and to phenotype and monitor broiler gait. These insights can be used to understand differences in the walking patterns of lame broilers, and to build more sophisticated gait prediction models.

When Everything Becomes Bigger: Big Data for Big Poultry Production

In future decades, the demand for poultry meat and eggs is predicted to considerably increase in pace with human population growth. Although this expansion clearly represents a remarkable opportunity for the sector, it conceals a multitude of challenges. Pollution and land erosion, competition for limited resources between animal and human nutrition, animal welfare concerns, limitations on the use of growth promoters and antimicrobial agents, and increasing risks and effects of animal infectious diseases and zoonoses are several topics that have received attention from authorities and the public. The increase in poultry production must be achieved mainly through optimization and increased efficiency. The increasing ability to generate large amounts of data ("big data") is pervasive in both modern society and the farming industry. Information accessibility-coupled with the availability of tools and computational power to store, share, integrate, and analyze data with automatic and flexible algorithms-offers an unprecedented opportunity to develop tools to maximize farm profitability, reduce socio-environmental impacts, and increase animal and human health and welfare. A detailed description of all topics and applications of big data analysis in poultry farming would be infeasible. Therefore, the present work briefly reviews the application of sensor technologies, such as optical, acoustic, and wearable sensors, as well as infrared thermal imaging and optical flow, to poultry farming. The principles and benefits of advanced statistical techniques, such as machine learning and deep learning, and their use in developing effective and reliable classification and prediction models to benefit the farming system, are also discussed. Finally, recent progress in pathogen genome sequencing and analysis is discussed, highlighting practical applications in epidemiological tracking, and reconstruction of microorganisms' population dynamics, evolution, and spread. The benefits of the objective evaluation of the effectiveness of applied control strategies are also considered. Although human-artificial intelligence collaborations in the livestock sector can be frightening because they require farmers and employees in the sector to adapt to new roles, challenges, and competencies-and because several unknowns, limitations, and open-ended questions are inevitable-their overall benefits appear to be far greater than their drawbacks. As more farms and companies connect to technology, artificial intelligence (AI) and sensing technologies will begin to play a greater role in identifying patterns and solutions to pressing problems in modern animal farming, thus providing remarkable production-based and commercial advantages. Moreover, the combination of diverse sources and types of data will also become fundamental for the development of predictive models able to anticipate, rather than merely detect, disease occurrence. The increasing availability of sensors, infrastructures, and tools for big data collection, storage, sharing, and analysis-together with the use of open standards and integration with pathogen molecular epidemiology-have the potential to address the major challenge of producing higher-quality, more healthful food on a larger scale in a more sustainable manner, thereby protecting ecosystems, preserving natural resources, and improving animal and human welfare and health.

EFSA Panel on Animal Health and Animal Welfare (AHAW Panel), Nielsen SS, Alvarez J, Bicout DJ, Calistri P, Canali E, Drewe JA, Garin‐Bastuji B, Gonzales Rojas JL, Schmidt CG, Herskin M, Michel V, Miranda Chueca MÁ, Padalino B, Roberts HC, Spoolder H, Stahl K, Viltrop A, Winckler C, Berg C, Edwards S, Knierim U, Riber A, Salamon A, Tiemann I, Fabris C, Manakidou A, Mosbach‐Schulz O, Van der Stede Y, Vitali M, Velarde A. Welfare of ducks, geese and quail on farm

This Scientific Opinion concerns the welfare of Domestic ducks (Anas platyrhynchos domesticus), Muscovy ducks (Cairina moschata domesticus) and their hybrids (Mule ducks), Domestic geese (Anser anser f. domesticus) and Japanese quail (Coturnix japonica) in relation to the rearing of breeders, birds for meat, Muscovy and Mule ducks and Domestic geese for foie gras and layer Japanese quail for egg production. The most common husbandry systems (HSs) in the European Union are described for each animal species and category. The following welfare consequences are described and assessed for each species: restriction of movement, injuries (bone lesions including fractures and dislocations, soft tissue lesions and integument damage and locomotory disorders including lameness), group stress, inability to perform comfort behaviour, inability to perform exploratory or foraging behaviour and inability to express maternal behaviour (related to prelaying and nesting behaviours). Animal-based measures relevant for the assessment of these welfare consequences were identified and described. The relevant hazards leading to the welfare consequences in the different HSs were identified. Specific factors such as space allowance (including minimum enclosure area and height) per bird, group size, floor quality, characteristics of nesting facilities and enrichment provided (including access to water to fulfil biological needs) were assessed in relation to the welfare consequences and, recommendations on how to prevent the welfare consequences were provided in a quantitative or qualitative way.

Individual Monitoring of Activity and Lameness in Conventional and Slower-Growing Breeds of Broiler Chickens Using Accelerometers

Accelerometers are increasingly being investigated to detect animal behavior as a method for monitoring individual welfare that overcomes manual challenges associated with time, resource, and discrete sampling. We investigated the effects of broiler chicken hybrid (hereafter breed) and weight on accelerometer activity (activityA; calculated as percentage of time spent active (%)) and its association with lameness as a major broiler welfare concern. Accelerometers were attached to birds of different breeds on between 2 and 4 occasions from 26 to 30 days old (conventional breed CNV) and 26 to 49 days old (two slower-growing breeds SGH; SGN). At 2.2 kg, lameness was scored using a 6-point gait scoring system (0: unaffected to 5: severely lame). Linear mixed effects models and breed-stratified generalized linear models together with a random-effect meta-analysis were used for data analyses. ActivityA was lower in faster-growing, heavier birds compared to slower-growing, lighter birds, showing overall consistency with previous behavioral research, but did not vary linearly with gait score. Accelerometers offer the potential for simple broad-scale continuous monitoring of broiler chicken activity behavior that requires limited data processing. Exploration of the ability of accelerometers to capture more subtle and specific changes in behavioral patterning, such as non-linear acceleration with gait score that could indicate early development of lameness, warrants further investigation.

Tracking and Characterizing Spatiotemporal and Three-Dimensional Locomotive Behaviors of Individual Broilers in the Three-Point Gait-Scoring System

Gait scoring is a useful measure for evaluating broiler production efficiency, welfare status, bone quality, and physiology. The research objective was to track and characterize spatiotemporal and three-dimensional locomotive behaviors of individual broilers with known gait scores by jointly using deep-learning algorithms, depth sensing, and image processing. Ross 708 broilers were placed on a platform specifically designed for gait scoring and manually categorized into one of three numerical scores. Normal and depth cameras were installed on the ceiling to capture top-view videos and images. Four birds from each of the three gait-score categories were randomly selected out of 70 total birds scored for video analysis. Bird moving trajectories and 16 locomotive-behavior metrics were extracted and analyzed via the developed deep-learning models. The trained model gained 100% accuracy and 3.62 ± 2.71 mm root-mean-square error for tracking and estimating a key point on the broiler back, indicating precise recognition performance. Broilers with lower gait scores (less difficulty walking) exhibited more obvious lateral body oscillation patterns, moved significantly or numerically faster, and covered more distance in each movement event than those with higher gait scores. In conclusion, the proposed method had acceptable performance for tracking broilers and was found to be a useful tool for characterizing individual broiler gait scores by differentiating between selected spatiotemporal and three-dimensional locomotive behaviors.

Perspectives of digital agriculture in diverse types of livestock supply chain systems. Making sense of uses and benefits

Digital technology is being introduced to global agriculture in a wide variety of forms that are collectively known as digital agriculture. In this paper we provide opportunities and value propositions of how this is occurring in livestock production systems, with a consistent emphasis on technology relating to animal health, animal welfare, and product quality for value creation. This is achieved by organizing individual accounts of digital agriculture in livestock systems according to four broad types-commodity-based; value seeking; subsistence and nature-based. Each type presents contrasting modes of value creation in downstream processing; as well as from the perspective of One Health. The ideal result of digital technology adoption is an equitable and substantial diversification of supply chains, increased monetization of animal product quality, and more sensitive management to meet customer demands and environmental threats. Such changes have a significance beyond the immediate value generated because they indicate endogenous growth in livestock systems, and may concern externalities imposed by the pursuit of purely commercial ends.

Seeing is caring – automated assessment of resource use of broilers with computer vision techniques

Routine monitoring of broiler chickens provides insights in the welfare status of a flock, helps to guarantee minimum defined levels of animal welfare and assists farmers in taking remedial measures at an early stage. Computer vision techniques offer exciting potential for routine and automated assessment of broiler welfare, providing an objective and biosecure alternative to the current more subjective and time-consuming methods. However, the current state-of-the-art computer vision solutions for assessing broiler welfare are not sufficient to allow the transition to fully automated monitoring in a commercial environment. Therefore, the aim of this study was to investigate the potential of computer vision algorithms for detection and resource use monitoring of broilers housed in both experimental and commercial settings, while also assessing the potential for scalability and resource-efficient implementation of such solutions. This study used a combination of detection and resource use monitoring methods, where broilers were first detected using Mask R-CNN and were then assigned to a specific resource zone using zone-based classifiers. Three detection models were proposed using different annotation datasets: model A with annotated broilers from a research facility, model B with annotated broilers from a commercial farm, and model A+B where annotations from both environments were combined. The algorithms developed for individual broiler detection performed well for both the research facility (model A, F1 score > 0.99) and commercial farm (model A+B, F1 score > 0.83) test data with an intersection over union of 0.75. The subsequent monitoring of resource use at the commercial farm using model A+B for broiler detection, also performed very well for the feeders, bale and perch (F1 score > 0.93), but not for the drinkers (F1 score = 0.28), which was likely caused by our evaluation method. Thus, the algorithms used in this study are a first step to measure resource use automatically in commercial application and allow detection of a large number of individual animals in a non-invasive manner. From location data of every frame, resource use can be calculated. Ultimately, the broiler detection and resource use monitoring might further be used to assess broiler welfare.

Differences and variation in welfare performance of broiler flocks in three production systems

There is a trend toward broiler production systems with higher welfare requirements, that use slower growing broiler strains, apply a reduced stocking density and provide environmental enrichment. Although these separate factors each contribute to increased broiler welfare, there is little information on their combined effect on broiler welfare under commercial conditions, and on the variation in welfare performance of flocks within production systems. The aim of this study was to compare the welfare performance and the between-flock variation in welfare of 3 Dutch commercial broiler production systems differing in welfare requirements: Conventional (C), Dutch Retail Broiler (DRB) and Better Life one star (BLS). We applied a welfare assessment method based on the Welfare Quality broiler assessment protocol, in which we used 5 animal-based welfare measures collected by slaughterhouses and hatcheries (mortality, footpad dermatitis, hock burn, breast irritation, scratches), and 3 resource- or management-based measures (stocking density, early feeding, environmental enrichment). Data were collected for at least 1889 flocks per production system over a 2-year period. To compare the different measures and to generate an overall flock welfare score, we calculated a score on a scale from 0 to 100 (bad-good) for each measure based on expert opinion. The overall flock score was the sum of the scores of the different welfare measures. The results showed that with increasing welfare requirements, a higher total welfare score was found across production systems (BLS > DRB > C; P < 0.0001). Regarding individual measures, C generally had lower (worse) scores than BLS and DRB (P < 0.05), except for scratches where C had highest (best) score (P < 0.001). Both welfare measure scores and the total welfare score of flocks showed large variation within and overlap between systems, and the latter especially when only the animal-based measures were included in the total flock score. Total flock score ranges including animal-based measures only were: 112.1 to 488.3 for C, 113.0 to 486.9 for DRB, 151.3 to 490.0 for BLS (on a scale from 0 [bad]–500 [good]), with median values of 330.8 for C, 370.9 for DRB, and 396.1 for BLS respectively. This indicates that factors such as farm management and day-old chick quality can have a major effect on the welfare performance of a flock and that there is room for welfare improvement in all production systems.

Bird Welfare in Zoos and Aquariums: General Insights across Industries

Animal welfare is a priority across accredited zoological institutions; however, historically, research has been prioritized for mammals. Bird-focused studies accounted for less than 10% of welfare research in zoos and aquariums over the last ten years. Due to the lack of scientific publications on bird welfare, zoo scientists and animal practitioners can look to other industries such as agriculture, laboratories, and companion animal research for insight. This qualitative review highlights findings across industries to inform animal care staff and scientists on the welfare needs of birds within zoos and aquariums. Specifically, the review includes an overview of research on different topics and a summary of key findings across nine resources that affect bird welfare. We also highlight areas where additional research is necessary. Future welfare research in zoos and aquariums should prioritize studies that consider a diversity of bird species across topics and work to identify animal-based measures with empirical evidence. Moving forward, research from other industries can help develop innovative research on bird welfare within zoos and aquariums.

The Relationships between Damaging Behaviours and Health in Laying Hens

Since the ban in January 2012 of conventional cages for egg production in the European Union (Council Directive 1999/74/EC), alternative systems such as floor, aviary, free-range, and organic systems have become increasingly common, reaching 50% of housing for hens in 2019. Despite the many advantages associated with non-cage systems, the shift to a housing system where laying hens are kept in larger groups and more complex environments has given rise to new challenges related to management, health, and welfare. This review examines the close relationships between damaging behaviours and health in modern husbandry systems for laying hens. These new housing conditions increase social interactions between animals. In cases of suboptimal rearing and/or housing and management conditions, damaging behaviour or infectious diseases are likely to spread to the whole flock. Additionally, health issues, and therefore stimulation of the immune system, may lead to the development of damaging behaviours, which in turn may result in impaired body conditions, leading to health and welfare issues. This raises the need to monitor both behaviour and health of laying hens in order to intervene as quickly as possible to preserve both the welfare and health of the animals.

In pursuit of a better broiler: walking ability and incidence of contact dermatitis in conventional and slower growing strains of broiler chickens

In this study, the mobility, incidence, and severity of contact dermatitis and litter moisture content were assessed in 14 strains of broiler chickens differing in growth rate. The strains encompassed 2 conventional (CONV; ADG0-48 > 60 g/d) and 12 slower growing (SG) strains categorized as FAST (ADG0-62 = 53-55 g/d), MOD (ADG0-62 = 50-51 g/d), and SLOW (ADG0-62 < 50 g/d), with 4 strains in each category. A total of 7,216 mixed-sex birds were equally allocated into 164 pens (44 birds/pen; 30 kg/m2) in a randomized incomplete block design, with each strain represented in 8 to 12 pens over 2-3 trials. From each pen, 4 to 6 birds were tested in the latency-to-lie (LTL) and group obstacle tests 1 wk prior to the birds reaching 2 target weights (TWs) of approximately 2.1 kg (TW1: 34 d for CONV and 48 d for SG strains) and 3.2 kg (TW2: 48 d for CONV and 62 d for SG strains). The incidence of footpad dermatitis (FPD) and hock burns (HB) were evaluated a day prior to each TW. Litter moisture content was determined biweekly from d 14 to d 56. At TW1, CONV and SLOW had longer LTL than FAST birds. At TW2, CONV, MOD, and FAST birds had similar LTL. At both TWs, CONV birds were lighter than FAST birds in the group obstacle test, yet their number of obstacle crossings was similar. At TW1, CONV birds had greater incidence of FPD than FAST and MOD, while at TW2, CONV birds had greater incidence than the other categories. The incidence of HB in CONV and MOD was greater than SLOW birds at TW1, while at TW2, the incidence of HB was greater in CONV and FAST birds vs. MOD and SLOW birds. Litter moisture content was high in all categories from d 28 onward. Our results indicate that both BW and growth rate influence leg strength and walking ability, whereas the overall high litter moisture content and to a lesser extent growth rate influenced the incidence of contact dermatitis.

Computer-Vision-Based Indexes for Analyzing Broiler Response to Rearing Environment: A Proof of Concept

Simple Summary

We tested two computer-vision-based indexes to analyze the rearing-environment enrichment on broiler movement as a function of comfort temperature and heat stress. The results indicated that the simultaneous application of cluster and unrest indexes could monitor the movement of the group of broilers under different environmental conditions. Future monitoring and alert systems based on computer vision should consider the complexity of the environment for detecting heat stress in broiler production.

Abstract

Computer-vision systems for herd detection and monitoring are increasingly present in precision livestock. This technology provides insights into how environmental variations affect the group’s movement pattern. We hypothesize that the cluster and unrest indexes based on computer vision (CV) can simultaneously assess the movement variation of reared broilers under different environmental conditions. The present study is a proof of principle and was carried out with twenty broilers (commercial strain Cobb^®), housed in a controlled-environment chamber. The birds were divided into two groups, one housed in an enriched environment and the control. Both groups were subjected to thermal comfort conditions and heat stress. Image analysis of individual or group behavior is the basis for generating animal-monitoring indexes, capable of creating real-time alert systems, predicting welfare, health, environment, and production status. The results obtained in the experiment in a controlled environment allowed the validation of the simultaneous application of cluster and unrest indexes by monitoring the movement of the group of broilers under different environmental conditions. Observational results also suggest that research in more significant proportions should be carried out to evaluate the potential positive impact of environmental enrichment in poultry production. The complexity of the environment is a factor to be considered in creating alert systems for detecting heat stress in broiler production. In large groups, birds’ movement and grouping patterns may differ; therefore, the CV system and indices will need to be recalibrated.

Laser Enrichment Device Stimulates Broiler Laser-Following Behavior While Increasing Individual Bird Locomotion and Pen-Wide Movement

Genetic selection for fast growth rate and high breast muscle yield in modern broilers has unintended effects on animal welfare and behavior, namely in terms of inactivity and leg disorders. We hypothesized that exercise stimulated through environmental enrichment could positively stimulate pen-wide activity and improve bird welfare. The study objectives were to implement a laser enrichment device to motivate active and feeding behaviors throughout the pen. Twelve hundred Ross 708 broilers were randomly assigned to enrichment (LASER; laser enrichment, or CON; no laser enrichment) for 49 d. Seventy focal birds were randomly assigned to 14 video-recorded pens for behavioral analysis, including focal bird home pen behavior and walking distance. Pen-wide activity was also measured during the 4-min laser periods, four times daily, d0–8, and 1 day weekly, wk 1–6. Focal birds were gait scored wk 1–6, and were euthanized on d42 for tibia bone mineral content, density, and bone breaking strength analysis. Time spent active was increased in LASER-enriched birds compared to CON on wk 3–5 by up to 214% (wk 4), and percent of time at the feeder was increased in LASER-enriched birds by 761% on wk 4 (P &lt; 0.05). Peak percent of birds following the laser (LASER-enriched pens only) was observed on d0 (8.52%). Over wk 1–6, peak laser-following behavior was observed on wk 3 (3.07% of birds). Percent of birds moving during laser periods was increased in LASER-enriched pens on d0, 1, 2, 6, 7, and 8, with a percent increase of 68.7% observed on d1 (P &lt; 0.05). Percent of birds moving (laser-following or not) was increased on wk 1, 3, and 4 in LASER-enriched pens, with an increase of 69.7% observed on wk 4 (P &lt; 0.05). No differences were found in tibia measures. These data indicate that laser enrichment stimulated voluntary locomotion through wk 5 and laser-following behavior through wk 6, and that the relatively small percent of birds actively following the laser stimulated pen-wide movement above the level of the CON through wk 4 on study.

Research on the Correlation between Breeding Environment and Activity of Yellow Feather Broilers Based on the Multichromatic Aberration Model

Broiler behavior is closely related to the breeding environment. Therefore, studying broiler behavior helps breeding farm workers to better carry out welfare breeding. In the breeding environment of yellow feather broilers, temperature, humidity, and ammonia concentration are the main factors that affect the behavior of the broilers. This study used a multichromatic aberration model to process the color images of yellow feather broilers to extract the activity feature of the broilers at different periods, utilized the Cb component of YCbCr color model and the b component of Lab color model to remove background litter in images, and employed the Q component of YIQ color model to remove the feeders and the drinkers from the image. The segmented images were constructed into an accumulator to generate a heat map of yellow feather broilers' activity. Then, the correlation between the activity and the temperature and humidity index (THI) and the correlation between the activity and ammonia concentration were explored. The experiment found that the activity of the broilers was significantly positively correlated with ammonia concentration (P < 0.05), indicating that the activity of yellow feather broilers increased with ammonia concentration ascending. Besides, the THI in the broiler chamber had a significant positive correlation with the ammonia data (P < 0.01), indicating that when the THI in the broiler chamber increases, the ammonia concentration rises. The research provides a direction for exploring the impact of THI and ammonia concentration on the performance of yellow feather broilers. At the same time, it provides a theoretical basis for the early warning and judgment of broiler breeding by farm workers in the future.

Movement Analysis to Associate Broiler Walking Ability with Gait Scoring

The genetic development of the commercial broiler has led to body misconfiguration and consequent walking disabilities, mainly at the slaughter age. The present study aimed to identify broiler locomotion ability using image analysis automatically. A total of 40 broilers that were 40 d old (male and female) were placed to walk on a specially built runway, and their locomotion was recorded. An image segmentation algorithm was developed, and the coordinates of the bird’s center of mass were extracted from the segmented images for each frame analyzed, and the unrest index (UI) was applied. We calculated the center of mass’s movement of the broiler walking lateral images capturing the bird’s displacement speed in the onward direction. Results indicated that broiler walking speed on the runway tends to decrease with the increase of the gait score. The locomotion did not differ between males or females. The proposed algorithm was efficient in predicting the broiler gait score based on their displacement speed.

Groups and Individuals: Optical Flow Patterns of Broiler Chicken Flocks Are Correlated with the Behavior of Individual Birds

Group level measures of welfare flocks have been criticized on the grounds that they give only average measures and overlook the welfare of individual animals. However, we here show that the group-level optical flow patterns made by broiler flocks can be used to deliver information not just about the flock averages but also about the proportion of individuals in different movement categories. Mean optical flow provides information about the average movement of the whole flock while the variance, skew and kurtosis quantify the variation between individuals. We correlated flock optical flow patterns with the behavior and welfare of a sample of 16 birds per flock in two runway tests and a water (latency-to-lie) test. In the runway tests, there was a positive correlation between the average time taken to complete the runway and the skew and kurtosis of optical flow on day 28 of flock life (on average slow individuals came from flocks with a high skew and kurtosis). In the water test, there was a positive correlation between the average length of time the birds remained standing and the mean and variance of flock optical flow (on average, the most mobile individuals came from flocks with the highest mean). Patterns at the flock level thus contain valuable information about the activity of different proportions of the individuals within a flock.

Symposium review: Precision technologies for dairy calves and management applications

There is an increasing interest in using precision dairy technologies (PDT) to monitor real-time animal behavior and physiology in livestock systems around the world. Although PDT in adult cattle is extensively reviewed, PDT use for the management of preweaned dairy calves has not been reviewed. We systematically reviewed research on the use and application of precision technologies in calves. Accelerometers have the potential to be used to monitor lying behavior, step activity, and rumination, which are useful to detect changes in behavior that may be indicative of disease, responses to painful procedures, or positive welfare behaviors such as play. Automated calf feeding systems can control delivery of nutritional plans to individualize feeding and weaning of calves; changes in feeding behaviors (such as milk intake, drinking speed, and unrewarded visits) may also be used to identify early onset of disease. The PDT devices also measure physiological and physical attributes in dairy calves. For instance, temperature monitoring devices such as infrared thermography, ruminal boluses, and implanted microchips have been assessed in calves, but no herd management-based commercial system is available. Many other PDT are in development with potential to be used in dairy calf management, such as image and acoustic-based monitoring, real-time location, and use of enrichment items for monitoring positive emotional states. We conclude that PDT have great potential for application in dairy calf management, enabling precise behavioral and physiological monitoring, targeted feeding programs, and identification of calves with poor health or behavioral impairments. We strongly encourage further development and validation of commercially available technologies for on-farm application of the monitoring of dairy calf welfare, performance, and health.

Estimation of the Gender Ratio of Chickens Based on Computer Vision: Dataset and Exploration

The gender ratio of free-range chickens is considered as a major animal welfare problem in commercial broiler farming. Free-range chicken producers need to identify chicken gender to estimate the economic value of their flock. However, it is challenging for farmers to estimate the gender ratio of chickens efficiently and accurately, since the environmental background is complicated and the chicken number is dynamic. Moreover, manual estimation is likely double counts or missed count and thus is inaccurate and time consuming. Hence, automated methods that can lead to results efficiently and accurately replace the identification abilities of a chicken gender expert, working in a farm environment, are beneficial to the industry. The contributions in this paper include: (1) Building the world’s first chicken gender classification database annotated manually, which comprises 800 chicken flock images captured on a farm and 1000 single chicken images separated from the flock images by an object detection network, labelled with gender information. (2) Training a rooster and hen classifier using a deep neural network and cross entropy in information theory to achieve an average accuracy of 96.85%. The evaluation of the algorithm performance indicates that the proposed automated method is practical for the gender classification of chickens on the farm environment and provides a feasible way of thinking for the estimation of the gender ratio.

Accuracy of Broiler Activity Index as Affected by Sampling Time Interval

Simple Summary

Broiler activity index is a measure of bird movement through determining bird-representative pixel changes between consecutive images. Since the concept of activity index was coined, different sampling time intervals of consecutive images have been used to determine broiler activity. In this study, we found that accuracy of broiler activity decreased at longer sampling time intervals, with the 0.04-s interval yielding the most accurate activity index among all intervals investigated. In addition, broiler activity in the commercial house generally decreased as birds aged and varied at different monitoring locations. The research provides insights into image-sampling strategies for accurately determining broiler activity index, which may help to address growing public concerns on poultry welfare and health.

Abstract

Different time intervals between consecutive images have been used to determine broiler activity index (AI). However, the accuracy of broiler AI as affected by sampling time interval remains to be explored. The objective of this study was to investigate the effect of the sampling time interval (0.04, 0.2, 1, 10, 60, and 300 s) on the accuracy of broiler AI at different bird ages (1–7 weeks), locations (feeder, drinker, and open areas) and times of day (06:00–07:00 h, 12:00–13:00 h, and 18:00–19:00 h). A ceiling-mounted camera was used to capture top-view videos for broiler AI calculations. The results show that the sampling time interval of 0.04 s yielded the highest broiler AI because more bird motion details were captured at this short time interval. The broiler AIs at longer time intervals were 1–99% of that determined at the 0.04-s interval. The broiler AI at 0.2-s interval showed an acceptable accuracy with 80% less computational resources. Broiler AI decreased as birds aged but increased after week 4 at the drinker area. Broiler AI was the highest at the open area for weeks 1–4 and at the feeder and drinker areas for weeks 5–7. It is concluded that the accuracy of broiler AI was significantly affected by sampling time intervals. Broiler AI in commercial housing showed both temporal and spatial variations.

A Machine Vision-Based Method for Monitoring Broiler Chicken Floor Distribution

The proper spatial distribution of chickens is an indication of a healthy flock. Routine inspections of broiler chicken floor distribution are done manually in commercial grow-out houses every day, which is labor intensive and time consuming. This task requires an efficient and automatic system that can monitor the chicken’s floor distributions. In the current study, a machine vision-based method was developed and tested in an experimental broiler house. For the new method to recognize bird distribution in the images, the pen floor was virtually defined/divided into drinking, feeding, and rest/exercise zones. As broiler chickens grew, the images collected each day were analyzed separately to avoid biases caused by changes of body weight/size over time. About 7000 chicken areas/profiles were extracted from images collected from 18 to 35 days of age to build a BP neural network model for floor distribution analysis, and another 200 images were used to validate the model. The results showed that the identification accuracies of bird distribution in the drinking and feeding zones were 0.9419 and 0.9544, respectively. The correlation coefficient (R), mean square error (MSE), and mean absolute error (MAE) of the BP model were 0.996, 0.038, and 0.178, respectively, in our analysis of broiler distribution. Missed detections were mainly caused by interference with the equipment (e.g., the feeder hanging chain and water line); studies are ongoing to address these issues. This study provides the basis for devising a real-time evaluation tool to detect broiler chicken floor distribution and behavior in commercial facilities.

Utilization of Optical Flow Algorithms to Monitor Development of Tail Biting Outbreaks in Pigs

A study was conducted to evaluate activity changes in pigs associated with the development of tail-biting outbreaks using optical flow algorithms. Pigs (n = 120; initial body weight = 25 ± 2.9 kg) housed in four pens of 30 pigs were studied for 13 weeks. Outbreaks of tail biting were registered through daily observations. Behavior of pigs in each pen was video-recorded. Three one-hour video segments, representing morning, noon, and afternoon on days 10, 7, and 3 before and during the first outbreak of tail biting were scanned at 5-min intervals to estimate time budget for lying, standing, eating, drinking, pig-directed behavior, and tail biting. The same video segments were analyzed for optical flow. Mean optical flow was higher three days before and during the tail-biting outbreak, compared to 10 days before the outbreak (p < 0.05), suggesting that pigs may increase their activity three days before tail-biting outbreaks. All optical flow measures (mean, variance, skewness, and kurtosis) were correlated (all p < 0.01) with time spent standing, indicating that movement during standing may be associated with optical flow measures. These results suggest that optical flow might be a promising tool for automatically monitoring activity changes to predict tail-biting outbreaks in pigs.

How Are Information Technologies Addressing Broiler Welfare? A Systematic Review Based on the Welfare Quality® Assessment

This systematic review aims to explore how information technologies (ITs) are currently used to monitor the welfare of broiler chickens. The question posed for the review was “which ITs are related to welfare and how do they monitor this for broilers?”. The Welfare Quality® (WQ) protocol for broiler assessment was utilized as a framework to analyse suitable articles. A total of 57 studies were reviewed wherein all principles of broiler welfare were addressed. The “good health” principle was the main criteria found to be addressed by ITs and IT-based studies (45.6% and 46.1%, respectively), whereas the least observed principle was “good feeding” (8.8%). This review also classified ITs and IT-based studies by their utilization (location, production system, variable measured, aspect of production, and experimental/practical use). The results show that the current focus of ITs is on problems with conventional production systems and that less attention has been given to free-range systems, slaughterhouses, and supply chain issues. Given the valuable results evidenced by the exploitation of ITs, their use in broiler production should continue to be encouraged with more attention given to farmer adoption strategies.

A novel environmental enrichment device increased physical activity and walking distance in broilers

Modern broilers are selected for fast growth and a large proportion of breast tissue, contributing to a top-heavy phenotype, leg disorders, and inactivity as birds reach market weight. Therefore, the objective was to motivate broilers to move through environmental enrichment. A total of 1,200 Ross 308 broilers were housed in pens of 30 for 6 wk: 600 birds were exposed to a novel laser enrichment device (LASER) and 600 were control. Each device projected 2 randomly moving red laser dots onto the floor 4 times/day for 4-min "laser periods." Seven LASER and 7 control pens, with 5 focal birds/pen (n = 70), were randomly selected to be video-recorded day 0 to 8 and once weekly for the remainder of the trial. Videos were analyzed to measure broiler time-budget and behaviors such as latency to feed and distance walked during laser periods. Focal birds were gait scored weekly on-farm. A test of the human-approach paradigm was carried out on weeks 1 and 6 on all pens. LASER birds were more active on days 0, 1, 3, 4, 5, 7, and 8, moving 254% more on day 7 (P ≤ 0.05). Time spent active was increased in LASER treatment by 114% on week 2; 157% on week 3; 90% on week 4; and 82% on week 5. LASER birds spent more time at the feeder on days 0, 1, 2, 5, 8, and on weeks 1 and 5, with 84% more time at feeder than control on day 5 (P ≤ 0.05). LASER birds walked further during laser periods on day 0 to 8, reaching 646.5 cm greater (day 1), and on weeks 2, 3, 4, and 5, with an increase of 367.5 cm on week 2 (P ≤ 0.05). Over week 1 to 6, 60.54 ± 7.4% of focal birds in the laser treatment were at the feeder during or within 5 min following laser periods. The laser enrichment device was successful in stimulating broiler physical activity and feeding, and did not negatively impact walking ability.

Recording behaviour of indoor-housed farm animals automatically using machine vision technology: A systematic review

Large-scale phenotyping of animal behaviour traits is time consuming and has led to increased demand for technologies that can automate these procedures. Automated tracking of animals has been successful in controlled laboratory settings, but recording from animals in large groups in highly variable farm settings presents challenges. The aim of this review is to provide a systematic overview of the advances that have occurred in automated, high throughput image detection of farm animal behavioural traits with welfare and production implications. Peer-reviewed publications written in English were reviewed systematically following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. After identification, screening, and assessment for eligibility, 108 publications met these specifications and were included for qualitative synthesis. Data collected from the papers included camera specifications, housing conditions, group size, algorithm details, procedures, and results. Most studies utilized standard digital colour video cameras for data collection, with increasing use of 3D cameras in papers published after 2013. Papers including pigs (across production stages) were the most common (n = 63). The most common behaviours recorded included activity level, area occupancy, aggression, gait scores, resource use, and posture. Our review revealed many overlaps in methods applied to analysing behaviour, and most studies started from scratch instead of building upon previous work. Training and validation sample sizes were generally small (mean±s.d. groups = 3.8±5.8) and in data collection and testing took place in relatively controlled environments. To advance our ability to automatically phenotype behaviour, future research should build upon existing knowledge and validate technology under commercial settings and publications should explicitly describe recording conditions in detail to allow studies to be reproduced.

Robots mediating interactions between animals for interspecies collective behaviors

Self-organized collective behavior has been analyzed in diverse types of gregarious animals. Such collective intelligence emerges from the synergy between individuals, which behave at their own time and spatial scales and without global rules. Recently, robots have been developed to collaborate with animal groups in the pursuit of better understanding their decision-making processes. These biohybrid systems make cooperative relationships between artificial systems and animals possible, which can yield new capabilities in the resulting mixed group. However, robots are currently tailor-made to successfully engage with one animal species at a time. This limits the possibilities of introducing distinct species-dependent perceptual capabilities and types of behaviors in the same system. Here, we show that robots socially integrated into animal groups of honeybees and zebrafish, each one located in a different city, allowing these two species to interact. This interspecific information transfer is demonstrated by collective decisions that emerge between the two autonomous robotic systems and the two animal groups. The robots enable this biohybrid system to function at any distance and operates in water and air with multiple sensorimotor properties across species barriers and ecosystems. These results demonstrate the feasibility of generating and controlling behavioral patterns in biohybrid groups of multiple species. Such interspecies connections between diverse robotic systems and animal species may open the door for new forms of artificial collective intelligence, where the unrivaled perceptual capabilities of the animals and their brains can be used to enhance autonomous decision-making, which could find applications in selective "rewiring" of ecosystems.

Associations among gait score, production data, abattoir registrations, and postmortem tibia measurements in broiler chickens

Lameness and impaired walking ability in rapidly growing meat-type broiler chickens are major welfare issues that cause economic losses. This study analyzed the prevalence of impaired walking and its associations with production data, abattoir registrations, and postmortem tibia measurements in Norwegian broiler chickens. Gait score (GS) was used to assess walking ability in 59 different commercial broiler flocks (Ross 308) close to the slaughter d, 5,900 broilers in total, in 3 different geographical regions. In each flock, 100 arbitrary broilers were gait scored and 10 random broilers were culled to harvest tibias. Abattoir registrations on flock level were collected after slaughter. A total of 24.6% of the broilers had moderate to severe gait impairment. The broilers were sampled in 2 stages, first slaughterhouse/region, and then owner/flock. The final models showed that impaired gait is associated with first-week mortality (P < 0.05), region (P < 0.001), height of tibias mid-shaft (P < 0.05), and calcium content in the tibia ash (P < 0.05), and negatively associated with DOA (P < 0.05). The prevalence of impaired gait indicates that this is a common problem in the broiler industry in Norway, although the mean slaughter age is only 31 d and the maximum allowed animal density is relatively low. Impaired walking ability could not be predicted by the welfare indicators footpad lesion score, total on-farm mortality, and decreasing DOA prevalence. Further studies are needed to explore the relationship between first-week mortality and gait score.

Genetic basis of leg health and its relationship with body weight in purebred turkey lines

The aims of this study were to estimate the genetic parameters for leg and foot health and mobility in purebred turkey lines and their genetic correlations with BW. Traits were gait score (GS) as an overall measure of leg health, footpad dermatitis (FPD), and 2 skeletal leg health traits, namely, valgus and varus deformities (VVD) and tibial dyschondroplasia (TD). Data from 4 different lines, comprising 3 yr of phenotypic records and 4 yr of pedigree information per line, were used. The sex average BW for the lines at 18 wk ranged from 19.1 kg (line A) to 12.4 kg (line D). The prevalence of VVD ranged from 5.2 to 14.6% and for TD from 4.1 to 23.2%. The average score for FPD on a scale of 0 to 100 ranged from 48.5 to 61.1. Gait Score was scored on a scale of 1 to 5, standardized to a mean of 3 and SD of 1. Heritabilities were estimated at 0.08 to 0.13 for GS, 0.01 to 0.07 for VVD, 0.06 to 0.12 for TD, and 0.10 to 0.15 for FPD (all SE ≤ 0.02). Estimates of the genetic correlations between VVD and TD ranged from 0.03 to 0.21 (all SE ≤ 0.08), and estimates of these with GS ranged from 0.07 to 0.87 (all SE ≤ 0.09). The genetic correlations of FPD with GS ranged from 0.00 to 0.34 (all SE ≤ 0.04), and with the skeletal leg health traits from -0.06 to 0.33 (all SE ≤ 0.06). Body weight showed estimated genetic correlations ranging from 0.28 to 0.51 (all SE ≤ 0.06) with GS, -0.06 to 0.50 (all SE ≤ 0.13) with VVD/TD and 0.05 to 0.34 (all SE ≤ 0.05) with FPD. The results suggest that selection for improved leg health can be incorporated effectively in a commercial turkey breeding program using balanced breeding goals, in which production traits and leg health traits are considered simultaneously.

Lameness assessment with automatic monitoring of activity in commercial broiler flocks

The possibility of using automatic recordings of broiler chicken activity in commercial flocks to assess the birds΄ walking ability (lameness) was investigated. Data were collected from 5 commercial broiler farms in 4 European countries, using 16 flocks and 33 assessment occasions. Lameness was assessed using established gait scoring methods (Kestin et al., 1992; Welfare Quality®, 2009) and took place at 3, 4, and 5 wk of age. Gait score (GS) was used to assess the birds' walking ability, and automatic recordings of bird activity were collected using the eYeNamic™ camera system before, during, and after an assessor walked through the house. The variables used to predict the level of GS extracted from the camera system were: baseline activity, time from assessor leaving the house to resumption of baseline activity, average activity over that period, and Δ Amplitude (difference between highest activity peak after assessor left the house and baseline level). Age (<0.001) and Δ Amplitude (P = 0.0002) were significantly related to GS, with the gait getting poorer with increased age and Δ Amplitude decreasing with declining walking ability. Both measures are thus included in a predictive equation. The results demonstrate a potential method using image analysis techniques to realize an automated assessment of the level of lameness in commercial broiler flocks. This could be of use in future animal welfare assessment schemes.

ANIMAL BEHAVIOR AND WELL-BEING SYMPOSIUM: Poultry welfare assessments: Current use and limitations

Recent attention has been given to developing welfare assessment tools for research purposes and for use directly on poultry farms. Historically, most of these tools have relied on resource- and management-based measures, but it is unclear how well they correlate with outcomes indicative of positive animal welfare. The subjective nature of many of these tools also makes it difficult to generalize across studies and farms without extensive training. More recently, the European Union Welfare Quality project set out to design assessment tools that were scientifically based and combined resource- and management-based measures with animal-based measures. Adding these measures was especially important for farm-level comparisons where farms may be utilizing different housing systems with inherent differences affecting the utility of resource- and management-based measures. The Welfare Quality Assessment protocol for poultry (WQA) offers researchers a tool that has been validated, tested for repeatability, and standardized across farms. This tool has been used in the United States and Canada at both the experimental and farm levels. The WQA is continuing to be developed, but many measures are still in need of validation and benchmarking as well as creating measures that can be used without needing extensive training to be successful on a go-forward basis. Tools must be designed for North American commercial production with a better understanding of the appropriate sample size, as well as their utility for use in alternative and extensive housing systems. The WQA shows promise in helping us to understand the influence of genetics, housing design, and management factors on the welfare of poultry.

Association between environmental predisposing risk factors and leg disorders in broiler chickens

Footpad dermatitis and lameness are a major welfare concern in broiler chicken farming. In general, footpad lesions are linked to poor environmental conditions. Ulcers that arise from advanced lesions can negatively affect the gait of the birds, with effects on the animal welfare, including, in the worst cases, inability to reach the feed or water. In this study, the degree of footpad dermatitis and lameness was manually scored on 4 broiler farms across Europe, as part of an EU-wide welfare assessment program. The welfare of the chickens was assessed 3 times per production cycle (at wk 3, 4, and 5), scoring footpad dermatitis, lameness, and litter quality. In the same broiler farms, variables such as air temperature and relative humidity were automatically measured over the same period. These variables were combined into a widely accepted thermal comfort index and associated to upper and lower thresholds, which made it possible to quantify the percentage of time the birds spent out of the thermal comfort zone (POOC). The data was analyzed by combining data from the welfare assessments with environmental data collected by the automated monitoring systems. Considering the comparison between POOC classes, the highest probabilities of footpad dermatitis and lameness were obtained when POOC values exceeded the 70% threshold. Therefore, the analysis showed that footpad dermatitis and lameness were more frequent when the flock was exposed to poor environmental conditions for prolonged periods ( < 0.001). Since environmental conditions can be continuously measured, and the risk factor for footpad dermatitis and lameness increases with poor environmental conditions, there is the possibility to develop a detection and control system of severe lesions.

Early warning of footpad dermatitis and hockburn in broiler chicken flocks using optical flow, bodyweight and water consumption

Footpad dermatitis and hockburn are serious welfare and economic issues for the production of broiler (meat) chickens. The authors here describe the use of an inexpensive camera system that monitors the movements of broiler flocks throughout their lives and suggest that it is possible to predict, even in young birds, the cross-sectional prevalence at slaughter of footpad dermatitis and hockburn before external signs are visible. The skew and kurtosis calculated from the authors' camera-based optical flow system had considerably more power to predict these outcomes in the 50 flocks reported here than water consumption, bodyweight or mortality and therefore have the potential to inform improved flock management through giving farmers early warning of welfare issues. Further trials are underway to establish the generality of the results.

Monitoring chicken flock behaviour provides early warning of infection by human pathogen Campylobacter

Campylobacter is the commonest bacterial cause of gastrointestinal infection in humans, and chicken meat is the major source of infection throughout the world. Strict and expensive on-farm biosecurity measures have been largely unsuccessful in controlling infection and are hampered by the time needed to analyse faecal samples, with the result that Campylobacter status is often known only after a flock has been processed. Our data demonstrate an alternative approach that monitors the behaviour of live chickens with cameras and analyses the 'optical flow' patterns made by flock movements. Campylobacter-free chicken flocks have higher mean and lower kurtosis of optical flow than those testing positive for Campylobacter by microbiological methods. We show that by monitoring behaviour in this way, flocks likely to become positive can be identified within the first 7-10 days of life, much earlier than conventional on-farm microbiological methods. This early warning has the potential to lead to a more targeted approach to Campylobacter control and also provides new insights into possible sources of infection that could transform the control of this globally important food-borne pathogen.

Animal welfare: a social networks perspective

Social network theory provides a useful tool to study complex social relationships in animals. The possibility to look beyond dyadic interactions by considering whole networks of social relationships allows researchers the opportunity to study social groups in more natural ways. As such, network-based analyses provide an informative way to investigate the factors influencing the social environment of group-living animals, and so has direct application to animal welfare. For example, animal groups in captivity are frequently disrupted by separations, reintroductions and/or mixing with unfamiliar individuals and this can lead to social stress and associated aggression. Social network analysis ofanimal groups can help identify the underlying causes of these socially-derived animal welfare concerns. In this review we discuss how this approach can be applied, and how it could be used to identify potential interventions and solutions in the area of animal welfare.

Injurious tail biting in pigs: how can it be controlled in existing systems without tail docking?

Tail biting is a serious animal welfare and economic problem in pig production. Tail docking, which reduces but does not eliminate tail biting, remains widespread. However, in the EU tail docking may not be used routinely, and some 'alternative' forms of pig production and certain countries do not allow tail docking at all. Against this background, using a novel approach focusing on research where tail injuries were quantified, we review the measures that can be used to control tail biting in pigs without tail docking. Using this strict criterion, there was good evidence that manipulable substrates and feeder space affect damaging tail biting. Only epidemiological evidence was available for effects of temperature and season, and the effect of stocking density was unclear. Studies suggest that group size has little effect, and the effects of nutrition, disease and breed require further investigation. The review identifies a number of knowledge gaps and promising avenues for future research into prevention and mitigation. We illustrate the diversity of hypotheses concerning how different proposed risk factors might increase tail biting through their effect on each other or on the proposed underlying processes of tail biting. A quantitative comparison of the efficacy of different methods of provision of manipulable materials, and a review of current practices in countries and assurance schemes where tail docking is banned, both suggest that daily provision of small quantities of destructible, manipulable natural materials can be of considerable benefit. Further comparative research is needed into materials, such as ropes, which are compatible with slatted floors. Also, materials which double as fuel for anaerobic digesters could be utilised. As well as optimising housing and management to reduce risk, it is important to detect and treat tail biting as soon as it occurs. Early warning signs before the first bloody tails appear, such as pigs holding their tails tucked under, could in future be automatically detected using precision livestock farming methods enabling earlier reaction and prevention of tail damage. However, there is a lack of scientific studies on how best to respond to outbreaks: the effectiveness of, for example, removing biters and/or bitten pigs, increasing enrichment, or applying substances to tails should be investigated. Finally, some breeding companies are exploring options for reducing the genetic propensity to tail bite. If these various approaches to reduce tail biting are implemented we propose that the need for tail docking will be reduced.

AUDIO-VISUAL RECOGNITION OF GOOSE FLOCKING BEHAVIOR

Every year, agriculture experience significant economic loss due to wild geese, rooks and other flocks of birds. A wide range of devices to detect and deter animals causing conflict is used to prevent this, although their effectiveness is often highly variable, due to habituation to disruptive or disturbing stimuli. Automated recognition of behaviors could form a critical component of a system capable of altering the disruptive stimulus to avoid habituation. This paper presents an audio-visual-based approach for recognition of goose flocking behavior. The vocal communication and movement of the flock is used for the audio-visual recognition, which is accomplished through classifier fusion of an acoustic and a video-based classifier. Acoustic behavior recognition is based on generalized perceptual features and support vector machines, and visual behavior recognition is based on optical flow estimation and a Bayesian Rule-Based scheme. Classifier fusion is implemented using the product rule on the soft-outputs from both classifiers. The algorithm has been used to recognize goose flocking behaviors (landing, foraging and flushing) and have improved the performance compared to using audio- or video-based classifiers alone. The improvement of using classifier fusion is most evident in the case of flushing and landing behavior recognition, where it was possible to combine the advantages of both the audio- and video-based classifier.

Modelling Farm Animal Welfare

The use of models in the life sciences has greatly expanded in scope and advanced in technique in recent decades. However, the range, type and complexity of models used in farm animal welfare is comparatively poor, despite the great scope for use of modeling in this field of research. In this paper, we review the different modeling approaches used in farm animal welfare science to date, discussing the types of questions they have been used to answer, the merits and problems associated with the method, and possible future applications of each technique. We find that the most frequently published types of model used in farm animal welfare are conceptual and assessment models; two types of model that are frequently (though not exclusively) based on expert opinion. Simulation, optimization, scenario, and systems modeling approaches are rarer in animal welfare, despite being commonly used in other related fields. Finally, common issues such as a lack of quantitative data to parameterize models, and model selection and validation are discussed throughout the review, with possible solutions and alternative approaches suggested.

Prediction of welfare outcomes for broiler chickens using Bayesian regression on continuous optical flow data

Currently, assessment of broiler (meat) chicken welfare relies largely on labour-intensive or post-mortem measures of welfare. We here describe a method for continuously and robustly monitoring the welfare of living birds while husbandry changes are still possible. We detail the application of Bayesian modelling to motion data derived from the output of cameras placed in commercial broiler houses. We show that the forecasts produced by the model can be used to accurately assess certain key aspects of the future health and welfare of a flock. The difference between healthy flocks and less-healthy ones becomes predictable days or even weeks before clinical symptoms become apparent. Hockburn (damaged leg skin, usually only seen in birds of two weeks or older) can be well predicted in flocks of only 1-2 days of age, using this approach. Our model combines optical flow descriptors of bird motion with robust multivariate forecasting and provides a sparse, efficient model with sparsity-inducing priors to achieve maximum predictive power with the minimum number of key variables.