Individualised automated lameness detection in dairy cows and the impact of historical window length on algorithm performance.

Lameness is an important economic problem in the dairy sector, resulting in production loss and reduced welfare of dairy cows. Given the modern-day expansion of dairy herds, a tool to automatically detect lameness in real-time can therefore create added value for the farmer. The challenge in developing camera-based tools is that one system has to work for all the animals on the farm despite each animal having its own individual lameness response. Individualising these systems based on animal-level historical data is a way to achieve accurate monitoring on farm scale. The goal of this study is to optimise a lameness monitoring algorithm based on back posture values derived from a camera for individual cows by tuning the deviation thresholds and the quantity of the historical data being used. Back posture values from a sample of 209 Holstein Friesian cows in a large herd of over 2000 cows were collected during 15 months on a commercial dairy farm in Sweden. A historical data set of back posture values was generated for each cow to calculate an individual healthy reference per cow. For a gold standard reference, manual scoring of lameness based on the Sprecher scale was carried out weekly by a single skilled observer during the final 6 weeks of data collection. Using an individual threshold, deviations from the healthy reference were identified with a specificity of 82.3%, a sensitivity of 79%, an accuracy of 82%, and a precision of 36.1% when the length of the healthy reference window was not limited. When the length of the healthy reference window was varied between 30 and 250 days, it was observed that algorithm performance was maximised with a reference window of 200 days. This paper presents a high-performing lameness detection system and demonstrates the importance of the historical window length for healthy reference calculation in order to ensure the use of meaningful historical data in deviation detection algorithms.

Review: Precision livestock farming: building 'digital representations' to bring the animals closer to the farmer.

Economic pressures continue to mount on modern-day livestock farmers, forcing them to increase herds sizes in order to be commercially viable. The natural consequence of this is to drive the farmer and the animal further apart. However, closer attention to the animal not only positively impacts animal welfare and health but can also increase the capacity of the farmer to achieve a more sustainable production. State-of-the-art precision livestock farming (PLF) technology is one such means of bringing the animals closer to the farmer in the facing of expanding systems. Contrary to some current opinions, it can offer an alternative philosophy to 'farming by numbers'. This review addresses the key technology-oriented approaches to monitor animals and demonstrates how image and sound analyses can be used to build 'digital representations' of animals by giving an overview of some of the core concepts of PLF tool development and value discovery during PLF implementation. The key to developing such a representation is by measuring important behaviours and events in the livestock buildings. The application of image and sound can realise more advanced applications and has enormous potential in the industry. In the end, the importance lies in the accuracy of the developed PLF applications in the commercial farming system as this will also make the farmer embrace the technological development and ensure progress within the PLF field in favour of the livestock animals and their well-being.

Effect of a photoperiodic green light programme during incubation on embryo development and hatch process.

This study was conducted to evaluate the effect of a 12-h light, 12-h dark (12L : 12D) photoperiod of green light during day 1 to day 18 of incubation time, on embryo growth, hormone concentration and the hatch process. In the test group, monochromatic light was provided by a total of 204 green light-emitting diodes (522 nm) mounted in a frame which was placed above the top tray of eggs to give even spread of illumination. No light-dark cycle was used in the control group. Four batches of eggs (n=300/group per batch) from fertile Ross 308 broiler breeders were used in this experiment. The beak length and crown-rump length of embryos incubated under green light were significantly longer than that of control embryos at day 10 and day 12, respectively (P<0.01). Furthermore, green light-exposed embryos had a longer third toe length compared with control embryos at day 10, day 14 and day 17 (P=0.02). At group level (n=4 batches), light stimulation had no effect on chick weight and quality at take-off, the initiation of hatch and hatch window. However, the individual hatching time of the light exposure focal chicks (n=33) was 3.4 h earlier (P=0.49) than the control focal chicks (n=36) probably due to the change in melatonin rhythm of the light group. The results of this study indicate that green light accelerates embryo development and alters hatch-related hormones (thyroid and corticosterone), which may result in earlier hatching.

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.

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.

Heart rate time series characteristics for early detection of infections in critically ill patients.

It is difficult to make a distinction between inflammation and infection. Therefore, new strategies are required to allow accurate detection of infection. Here, we hypothesize that we can distinguish infected from non-infected ICU patients based on dynamic features of serum cytokine concentrations and heart rate time series. Serum cytokine profiles and heart rate time series of 39 patients were available for this study. The serum concentration of ten cytokines were measured using blood sampled every 10 min between 2100 and 0600 hours. Heart rate was recorded every minute. Ten metrics were used to extract features from these time series to obtain an accurate classification of infected patients. The predictive power of the metrics derived from the heart rate time series was investigated using decision tree analysis. Finally, logistic regression methods were used to examine whether classification performance improved with inclusion of features derived from the cytokine time series. The AUC of a decision tree based on two heart rate features was 0.88. The model had good calibration with 0.09 Hosmer-Lemeshow p value. There was no significant additional value of adding static cytokine levels or cytokine time series information to the generated decision tree model. The results suggest that heart rate is a better marker for infection than information captured by cytokine time series when the exact stage of infection is not known. The predictive value of (expensive) biomarkers should always be weighed against the routinely monitored data, and such biomarkers have to demonstrate added value.

A decision-tree model to detect post-calving diseases based on rumination, activity, milk yield, BW and voluntary visits to the milking robot.

Early detection of post-calving health problems is critical for dairy operations. Separating sick cows from the herd is important, especially in robotic-milking dairy farms, where searching for a sick cow can disturb the other cows' routine. The objectives of this study were to develop and apply a behaviour- and performance-based health-detection model to post-calving cows in a robotic-milking dairy farm, with the aim of detecting sick cows based on available commercial sensors. The study was conducted in an Israeli robotic-milking dairy farm with 250 Israeli-Holstein cows. All cows were equipped with rumination- and neck-activity sensors. Milk yield, visits to the milking robot and BW were recorded in the milking robot. A decision-tree model was developed on a calibration data set (historical data of the 10 months before the study) and was validated on the new data set. The decision model generated a probability of being sick for each cow. The model was applied once a week just before the veterinarian performed the weekly routine post-calving health check. The veterinarian's diagnosis served as a binary reference for the model (healthy-sick). The overall accuracy of the model was 78%, with a specificity of 87% and a sensitivity of 69%, suggesting its practical value.

Lameness detection in dairy cattle: single predictor v. multivariate analysis of image-based posture processing and behaviour and performance sensing.

The objective of this study was to evaluate if a multi-sensor system (milk, activity, body posture) was a better classifier for lameness than the single-sensor-based detection models. Between September 2013 and August 2014, 3629 cow observations were collected on a commercial dairy farm in Belgium. Human locomotion scoring was used as reference for the model development and evaluation. Cow behaviour and performance was measured with existing sensors that were already present at the farm. A prototype of three-dimensional-based video recording system was used to quantify automatically the back posture of a cow. For the single predictor comparisons, a receiver operating characteristics curve was made. For the multivariate detection models, logistic regression and generalized linear mixed models (GLMM) were developed. The best lameness classification model was obtained by the multi-sensor analysis (area under the receiver operating characteristics curve (AUC)=0.757±0.029), containing a combination of milk and milking variables, activity and gait and posture variables from videos. Second, the multivariate video-based system (AUC=0.732±0.011) performed better than the multivariate milk sensors (AUC=0.604±0.026) and the multivariate behaviour sensors (AUC=0.633±0.018). The video-based system performed better than the combined behaviour and performance-based detection model (AUC=0.669±0.028), indicating that it is worthwhile to consider a video-based lameness detection system, regardless the presence of other existing sensors in the farm. The results suggest that Θ2, the feature variable for the back curvature around the hip joints, with an AUC of 0.719 is the best single predictor variable for lameness detection based on locomotion scoring. In general, this study showed that the video-based back posture monitoring system is outperforming the behaviour and performance sensing techniques for locomotion scoring-based lameness detection. A GLMM with seven specific variables (walking speed, back posture measurement, daytime activity, milk yield, lactation stage, milk peak flow rate and milk peak conductivity) is the best combination of variables for lameness classification. The accuracy on four-level lameness classification was 60.3%. The accuracy improved to 79.8% for binary lameness classification. The binary GLMM obtained a sensitivity of 68.5% and a specificity of 87.6%, which both exceed the sensitivity (52.1%±4.7%) and specificity (83.2%±2.3%) of the multi-sensor logistic regression model. This shows that the repeated measures analysis in the GLMM, taking into account the individual history of the animal, outperforms the classification when thresholds based on herd level (a statistical population) are used.

Higher levels of CO2 during late incubation alter the hatch time of chicken embryos.

1. It has been reported that the increasing CO2 tension triggers the embryo to pip the air cell and emerge from the egg. However, the mechanism by which higher CO2 concentrations during the last few days of incubation affect chick physiology and the hatching process is unclear. This study investigated the effect of CO2 concentrations up to 1% during pipping, on the onset and length of the hatch window (HW) and chick quality. 2. Four batches of Ross 308 broiler eggs (600 eggs per batch) were incubated in two small-scale custom-built incubators (Petersime NV). During the final 3 d of incubation, control eggs were exposed to a lower CO2 concentration (0.3%), while the test eggs experienced a higher CO2 concentration programme (peak of 1%). 3. There were no significant differences in blood values, organ weight and body weight. There was also no difference in hatchability between control and test groups. However, a small increase in the chick weight and the percentage of first class chicks was found in the test groups. Furthermore, plasma corticosterone profiles during hatching were altered in embryos exposed to higher CO2; however, they dropped to normal levels at d 21 of incubation. Importantly, the hatching process was delayed and synchronised in the test group, resulting in a narrowed HW which was 2.7 h shorter and 5.3 h later than the control group. 4. These results showed that exposing chicks to 1% CO2 concentration during pipping did not have negative impacts on physiological status of newly hatched chicks. In addition, it may have a significant impact on the physiological mechanisms controlling hatching and have benefits for the health and welfare of chickens by reducing the waiting time after hatching.

Discerning pig screams in production environments.

Pig vocalisations convey information about their current state of health and welfare. Continuously monitoring these vocalisations can provide useful information for the farmer. For instance, pig screams can indicate stressful situations. When monitoring screams, other sounds can interfere with scream detection. Therefore, identifying screams from other sounds is essential. The objective of this study was to understand which sound features define a scream. Therefore, a method to detect screams based on sound features with physical meaning and explicit rules was developed. To achieve this, 7 hours of labelled data from 24 pigs was used. The developed detection method attained 72% sensitivity, 91% specificity and 83% precision. As a result, the detection method showed that screams contain the following features discerning them from other sounds: a formant structure, adequate power, high frequency content, sufficient variability and duration.

An insight into the heat and mass transfer mechanisms of eggshells hatching broiler chicks and its effects to the hatcher environment.

Thermodynamic study of incubated eggs is an important component in the optimisation of incubation processes. However, research on the interaction of heat and moisture transfer mechanisms in eggs is rather limited and does not focus on the hatching stage of incubation. During hatch, both the recently hatched chick and the broken eggshell add extra heat and moisture contents to the hatcher environment. In this study, we have proposed a novel way to estimate thermodynamically the amount of water evaporated from a broken eggshell during hatch. The hypothesis of this study considers that previously reported drops in eggshell temperature during hatching of chicks is the result remaining water content evaporating from the eggshell, released on the inner membrane by the recently hatched wet chick, just before hatch. To reproduce this process, water was sprayed on eggshells to mimic the water-fluid from the wet body of a chick. For each sample of eggshell, the shell geometry and weight, surface area and eggshell temperature were measured. Water evaporation losses and convection coefficient were calculated using a novel model approach considering the simultaneous heat and mass transfer profiles in an eggshell. The calculated average convective coefficient was 23.9 ± 7.5 W/m(2) °C, similar to previously reported coefficients in literature as a function of 0.5-1m/s air speed range. Comparison between measured and calculated values for the water evaporation showed 68% probability accuracy, associated to the use of an experimentally derived single heat transfer coefficient. The results support our proposed modelling approach of heat and mass transfer mechanisms. Furthermore, by estimating the amount of evaporated water in an eggshell post-hatch, air humidity levels inside the hatcher can be optimised to ensure wet chicks dry properly while not dehydrating early hatching chicks.

Physiological status of broiler chicks at pulling time and the relationship to duration of holding period.

Newly hatched chicks may be held longer than 48 h and experience long periods of fasting in commercial hatcheries. Limited information is known about the physiological status of chicks in such situations, due to the difficulty of precisely recording time of hatch. This study investigated the effect of the time from hatch to pulling (holding period) on physiological measures/parameters in 109 broiler chicks. Fertile Ross 308 eggs were incubated in a custom built small-scale incubator. The individual hatching time of each focal chick was determined using eggshell temperature monitoring. At 'pulling' (512 h of incubation time), the quality of focal chicks was assessed using the chick scoring method and physiological parameters were measured including BW, organ (heart, liver and stomach) weights, blood values and plasma corticosterone level. The time from hatch to pulling varied from 7.58 to 44.97 h. Egg weight at setting was significantly correlated with chick BW and weight of organs at pulling, but had no effect on chick quality, blood values and plasma corticosterone. Relative BW at pulling was negatively associated with the duration of holding period (P=0.002). However, there was a positive correlation between relative stomach weight and the duration of the holding period (P<0.001). As the holding period duration increased, there was a trend that blood partial pressure of oxygen, haematocrit and haemoglobin also increased, and blood partial pressure of carbon dioxide, total carbon dioxide and bicarbonate decreased (P<0.05). A wide range of plasma corticosterone was observed from chicks that had experienced different durations of holding period. We conclude that shortening the hatch window and minimising the number of chicks that experience a long holding period before pulling may improve chick quality and physiological status, which may be due to unfavourable environmental conditions that include feed and water deprivation.

Effect of species-specific sound stimulation on the development and hatching of broiler chicks.

1. Previous research has reported that chicken embryos develop a functionary auditory system during incubation and that prenatal sound may play an important role in embryo development and alter the hatch time. In this study the effects of prenatal auditory stimulation on hatch process, hatch performance, the development of embryo and blood parameters were investigated. 2. Four batches of Ross 308 broiler breeder eggs were incubated either in control or in sound-stimulated groups. The sound-stimulated embryos were exposed to a discontinuous sound of species-specific calls by means of a speaker at 72 dB for 16 h a day: maternal calls from d 10 to d 19 of incubation time and embryo/chick calls from d 19 until hatching. The species-specific sound was excluded from the control group. 3. The onset of hatch was delayed in the sound-stimulated group compared to the controls. This was also supported by comparison of the exact hatching time of individual focal chicks within the two groups. However, the sound-stimulated embryos had a lower hatchability than the control group, mainly due to significantly increased numbers of late deaths. 4. The embryos exhibited a similar growth pattern between the sound-stimulated group and the control group. Although sound exposure decreased body weight at d 16, no consistent effect of sound on body weight at incubation stage was observed. Species-specific sound stimulation also had no impact on chick quality, blood values and plasma corticosterone concentrations during hatch.

Precision livestock farming technologies for welfare management in intensive livestock systems.

The worldwide demand for meat and animal products is expected to increase by at least 40% in the next 15 years. The first question is how to achieve high-quality, sustainable and safe meat production that can meet this demand. At the same time, livestock production is currently facing serious problems. Concerns about animal health in relation to food safety and human health are increasing. The European Union wants improved animal welfare and has made a significant investment in it. At the same time, the environmental impact of the livestock sector is a major issue. Finally, it is necessary to ask how the farmer, who is the central figure in this process, will make a living from more sustainable livestock production systems. One tool that might provide real opportunities is precision livestock farming (PLF). In contrast to previous approaches, PLF systems aim to offer a real-time monitoring and management system that focuses on improving the life of the animals by warning when problems arise so that the farmer may take immediate action. Continuous, fully automatic monitoring and improvement of animal health and welfare, product yields and environmental impacts should become possible. This paper presents examples of systems that have already been developed in order to demonstrate the potential benefits of this technology.

Different stressors elicit different responses in the salivary biomarkers cortisol, haptoglobin, and chromogranin A in pigs.

Most commonly, salivary cortisol is used in pig stress assessment, alternative salivary biomarkers are scarcely studied. Here, salivary cortisol and two alternative salivary biomarkers, haptoglobin and chromogranin A were measured in a pig stress study. Treatment pigs (n = 24) were exposed to mixing and feed deprivation, in two trials, and compared to untreated controls (n = 24). Haptoglobin differed for feed deprivation vs control. Other differences were only found within treatment. Treatment pigs had higher salivary cortisol concentrations on the mixing day (P < 0.05). Chromogranin A concentrations were increased on the day of refeeding (P < 0.05). Haptoglobin showed a similar pattern to chromogranin A. Overall correlations between the salivary biomarkers were positive. Cortisol and chromogranin A were moderately correlated (r = 0.49, P < 0.0001), correlations between other markers were weaker. The present results indicate that different types of stressors elicited different physiological stress responses in the pigs, and therefore including various salivary biomarkers in stress evaluation seems useful.

The effect of routine hoof trimming on locomotion score, ruminating time, activity, and milk yield of dairy cows.

The objective of this study was to quantify the effect of hoof trimming on cow behavior (ruminating time, activity, and locomotion score) and performance (milk yield) over time. Data were gathered from a commercial dairy farm in Israel where routine hoof trimming is done by a trained hoof trimmer twice per year on the entire herd. In total, 288 cows spread over 6 groups with varying production levels were used for the analysis. Cow behavior was measured continuously with a commercial neck activity logger and a ruminating time logger (HR-Tag, SCR Engineers Ltd., Netanya, Israel). Milk yield was recorded during each milking session with a commercial milk flow sensor (Free Flow, SCR Engineers Ltd.). A trained observer assigned on the spot 5-point locomotion scores during 19 nighttime milking occasions between 22 October 2012 and 4 February 2013. Behavioral and performance data were gathered from 1wk before hoof trimming until 1wk after hoof trimming. A generalized linear mixed model was used to statistically test all main and interactive effects of hoof trimming, parity, lactation stage, and hoof lesion presence on ruminating time, neck activity, milk yield, and locomotion score. The results on locomotion scores show that the proportional distribution of cows in the different locomotion score classes changes significantly after trimming. The proportion of cows with a locomotion score ≥3 increases from 14% before to 34% directly after the hoof trimming. Two months after the trimming, the number of cows with a locomotion score ≥3 reduced to 20%, which was still higher than the baseline values 2wk before the trimming. The neck activity level was significantly reduced 1d after trimming (380±6 bits/d) compared with before trimming (389±6 bits/d). Each one-unit increase in locomotion score reduced cow activity level by 4.488 bits/d. The effect of hoof trimming on ruminating time was affected by an interaction effect with parity. The effect of hoof trimming on locomotion scores was affected by an interaction effect with lactation stage and tended to be affected by interaction effects with hoof lesion presence, indicating that cows with a lesion reacted different to the trimming than cows without a lesion did. The results show that the routine hoof trimming affected dairy cow behavior and performance in this farm.

Effect of transportation duration of 1-day-old chicks on postplacement production performances and pododermatitis of broilers up to slaughter age.

This experiment studied the effect of transportation duration of 1-d-old chicks on dehydration, mortality, production performance, and pododermatitis during the growout period. Eggs from the same breeder flock (Ross PM3) were collected at 35, 45, and 56 wk of age, for 3 successive identical experiments. In each experiment, newly hatched chicks received 1 of 3 transportation duration treatments from the hatchery before placement in the on-site rearing facility: no transportation corresponding to direct placement in less than 5 min (T00), or 4 (T04) or 10 h (T10) of transportation. The chicks were housed in 35-m(2) pens (650 birds each) and reared until 35 d old. Hematocrit and chick BW were measured on sample chicks before and after transportation. During the growout period, bird weight, feed uptake, and feed conversion ratio were measured weekly until slaughter. Transportation duration affected BW; T00 groups had a significantly higher BW than T04 and T10 transported birds but this effect lasted only until d 21. No clear effect on hematocrit, feed uptake, feed conversion ratio, or mortality was observed for birds transported up to 10 h. The decrease in weight in T10 birds was associated with less severe pododermatitis. Increasing age of the breeder flock was correlated with reduced egg fertility and hatchability, and also with higher quality and BW of hatched chicks. Chicks from older breeders also exhibited reduced mortality during the growout period.

Lameness detection based on multivariate continuous sensing of milk yield, rumination, and neck activity.

The objective of this study was to develop and validate a mathematical model to detect clinical lameness based on existing sensor data that relate to the behavior and performance of cows in a commercial dairy farm. Identification of lame (44) and not lame (74) cows in the database was done based on the farm's daily herd health reports. All cows were equipped with a behavior sensor that measured neck activity and ruminating time. The cow's performance was measured with a milk yield meter in the milking parlor. In total, 38 model input variables were constructed from the sensor data comprising absolute values, relative values, daily standard deviations, slope coefficients, daytime and nighttime periods, variables related to individual temperament, and milk session-related variables. A lame group, cows recognized and treated for lameness, to not lame group comparison of daily data was done. Correlations between the dichotomous output variable (lame or not lame) and the model input variables were made. The highest correlation coefficient was obtained for the milk yield variable (rMY=0.45). In addition, a logistic regression model was developed based on the 7 highest correlated model input variables (the daily milk yield 4d before diagnosis; the slope coefficient of the daily milk yield 4d before diagnosis; the nighttime to daytime neck activity ratio 6d before diagnosis; the milk yield week difference ratio 4d before diagnosis; the milk yield week difference 4d before diagnosis; the neck activity level during the daytime 7d before diagnosis; the ruminating time during nighttime 6d before diagnosis). After a 10-fold cross-validation, the model obtained a sensitivity of 0.89 and a specificity of 0.85, with a correct classification rate of 0.86 when based on the averaged 10-fold model coefficients. This study demonstrates that existing farm data initially used for other purposes, such as heat detection, can be exploited for the automated detection of clinically lame animals on a daily basis as well.

Embryonic development and the physiological factors that coordinate hatching in domestic chickens.

Embryonic growth and development is influenced by both endogenous and exogenous factors. The purpose of this review is to discuss the critical stages of chick embryonic development in relation to functional maturation of numerous organ systems, the acquisition of thermoregulation, and the hatching process. In addition, the mechanism of hatching, including sound synchronization and hormonal and environmental stimulation, will be discussed. Finally, the importance of effective hatching synchronization mechanisms will also be highlighted.