Review: Exploring the use of precision livestock farming for small ruminant welfare management.

Small ruminant (sheep and goat) production of meat and milk is undertaken in diverse topographical and climatic environments and the systems range from extensive to intensive. This could lead to different types of welfare compromise, which need to be managed. Implementing Precision Livestock Farming (PLF) and other new or innovative technologies could help to manage or monitor animal welfare. This paper explores such opportunities, seeking to identify promising aspects of PLF that may allow improved management of welfare for small ruminants using literature search (two reviews), workshops in nine countries (France, Greece, Ireland, Israel, Italy, Norway, Romania, Spain, and the United Kingdom) with 254 stakeholders, and panels with 52 experts. An investigation of the main welfare challenges that may affect sheep and goats across the different management systems in Europe was undertaken, followed by a prioritisation of animal welfare issues obtained in the nine countries. This suggested that disease and health issues, feed access and undernutrition/malnutrition, maternal behaviour/offspring losses, environmental stressors and issues with agonistic behavioural interactions were important welfare concerns. These welfare issues and their indicators (37 for sheep, 25 for goats) were categorised into four broad welfare indicator categories: weight loss or change in body state (BWC), behavioural change (BC), milk yield and quality (MY), and environmental indicators (Evt). In parallel, 24 potential PLF and innovative technologies (8 for BWC; 10 for BC; 4 for MY; 6 for Evt) that could be relevant to monitor these broad welfare indicator categories and provide novel approaches to manage and monitor welfare have been identified. Some technologies had the capacity to monitor more than one broad indicator. Out of the 24 technologies, only 12 were animal-based sensors, or that could monitor the animal individually. One alternative could be to incorporate a risk management approach to welfare, using aspects of environmental stress. This could provide an early warning system for the potential risks of animal welfare compromise and alert farmers to the need to implement mitigation actions.

Biometric identification of dairy cows via real-time facial recognition.

Biometrics methods, which currently identify humans, can potentially identify dairy cows. Given that animal movements cannot be easily controlled, identification accuracy and system robustness are challenging when deploying an animal biometrics recognition system on a real farm. Our proposed method performs multiple-cow face detection and face classification from videos by adjusting recent state-of-the-art deep-learning methods. As part of this study, a system was designed and installed at four meters above a feeding zone at the Volcani Institute's dairy farm. Two datasets were acquired and annotated, one for facial detection and the second for facial classification of 77 cows. We achieved for facial detection a mean average precision (at Intersection over Union of 0.5) of 97.8% using the YOLOv5 algorithm, and facial classification accuracy of 96.3% using a Vision-Transformer model with a unique loss-function borrowed from human facial recognition. Our combined system can process video frames with 10 cows' faces, localize their faces, and correctly classify their identities in less than 20 ms per frame. Thus, up to 50 frames per second video files can be processed with our system in real-time at a dairy farm. Our method efficiently performs real-time facial detection and recognition on multiple cow faces using deep neural networks, achieving a high precision in real-time operation. These qualities can make the proposed system a valuable tool for an automatic biometric cow recognition on farms.

Machine vision-based automatic lamb identification and drinking activity in a commercial farm.

Using ear tags, farmers can track specific data for individual lambs such as age, medical records, body condition scores, genetic abnormalities; to make data-based decisions. However, automatic reading of ear tags using Radio Frequency Identification requires (a) an antenna, (b) a reader, (c) comparable reading standards; consequently, such a system can be expensive and impractical for a large group of lambs, especially in situations where animals are not required to have a compulsory Electronic identification, contrary to the case in Europe, where it is mandatory. Therefore, this paper proposes a machine vision system for indoor animals to identify individual lambs using existing ear tags. Using a camera that is installed such that the trough is visible, the drinking behaviour of the lambs can be automatically monitored. Data from different lamb groups in two different pens were collected. The identification algorithm includes a number of steps: (1) Detecting the lambs' face, and its ear tags in each image; (2) Cropping each ear tag image and discerning the digits on it to obtain the tag number; (3) Tracking each lamb throughout the visit using a tracking algorithm; (4) Recovering the ear tag number using an algorithm that incorporates a list of the ear tag numbers of the lambs in each pen, and the predictions for each lamb in each frame. The You Only Look Once deep learning object detection algorithm was applied to locate and localise the lamb's face and the digits in an image. The models' datasets contained 1 160 and 2 165 images for the training set, and 325 and 616 images for the validation set, respectively. The algorithm output includes the identity of each lamb that came to drink, and its duration. The identification system resulted in a total accuracy of 93% for the data tested, which consisted of approximately 900 visits to the drinking stations, and was collected in real time in a natural environment. The ground truth of each video of a visit was obtained by human observation by studying the video. We checked if there was indeed a visit to the water trough and if so we registered the ear tag number of each lamb whose head was above the water trough. Thus, identifying lambs in a commercial pen using a relatively inexpensive and easily installed system consisting of a RGB camera and a computer vision-based algorithm has potential for farm management.

A machine vision system to predict individual cow feed intake of different feeds in a cowshed.

Data on individual feed intake of dairy cows, an important variable for farm management, are currently unavailable in commercial dairies. A real-time machine vision system including models that are able to adapt to multiple types of feed was developed to predict individual feed intake of dairy cows. Using a Red-Green-Blue-Depth (RGBD) camera, images of feed piles of two different feed types (lactating cows' feed and heifers' feed) were acquired in a research dairy farm, for a range of feed weights under varied configurations and illuminations. Several models were developed to predict individual feed intake: two Transfer Learning (TL) models based on Convolutional Neural Networks (CNNs), one CNN model trained on both feed types, and one Multilayer Perceptron and Convolutional Neural Network model trained on both feed types, along with categorical data. We also implemented a statistical method to compare these four models using a Linear Mixed Model and a Generalised Linear Mixed Model, showing that all models are significantly different. The TL models performed best and were trained on both feeds with TL methods. These models achieved Mean Absolute Errors (MAEs) of 0.12 and 0.13 kg per meal with RMSE of 0.18 and 0.17 kg per meal for the two different feeds, when tested on varied data collected manually in a cowshed. Testing the model with actual cows' meals data automatically collected by the system in the cowshed resulted in a MAE of 0.14 kg per meal and RMSE of 0.19 kg per meal. These results suggest the potential of measuring individual feed intake of dairy cows in a cowshed using RGBD cameras and Deep Learning models that can be applied and tuned to different types of feed.

Short Communication: The effect of age on young sheep biometric identification.

Biometric identification provides an important tool for precision livestock farming. This study investigates the effect of weight gain and sheep maturation on recognition performance. Sheep facial identification was implemented using two convolutional neural network (CNN) called Faster R-CNN, and ResNet50V2, equipped with the state-of-art Additive Angular Margin (ArcFace) loss function. The identification model was tested on 47 young sheep at different stages, during a 3-month growth period, when they were between 2 and 5 months old, throughout which the sheep gained approximately 30 kilograms in weight. Results revealed that when the model was trained and tested on images of sheep aged 2 months, the average accuracy of the group was 95.4%, compared with 91.3% when trained on images of sheep aged 2 months but tested on images of sheep aged 5 months.

Design a system for measuring individual cow feed intake in commercial dairies.

Monitoring individual cow feed intake is necessary for calculating the cow individual feed efficiency. The cost and maintenance time necessary for research systems make them impractical for most of the commercial producers. We developed a measurement system with producer convenience and low investment as key design criteria. The goal of this study was to design the system and validate its ability to rank cows by their feed conversion efficiency in commercial farms. The new system consisted of three principal parts: (a) a hanging weighing system, (b) a visual cow identification system and (c) an automatic cleaning system. The weighing system consisted of hanging a single load cell to provide feed mass measurements. The image-based cow identification system (replacing Radio-Frequency Identification) entailed cameras installed above the feeding area and an image processing algorithm that recognized cows by their collar numbers. The new system worked within normal farm routines: the feed supplying truck distributed the animal feed, and a tractor cleaned feed residual. To validate the accuracy and convenience of the system and to rank the cows by their efficiency, an experiment with six scales and 12 cows was conducted in a research barn, succeeded by eight-scale system in a commercial farm with 16 cows. The feed intake of each cow participating in the experiments was monitored for one month. The validation experiment showed that the system had the following specification: scales were accurate within 120 g; the visual cow identification rate was greater than 96%; feeding duration was accurate to 52 s; and routine farm practices (feed distribution, pushing, and residual removal) continued as usual. The cost for a feeding station (utilized consequently for a number of cows) was about 1 500 USD. An example of application of the system to rank cows by their efficiency under commercial conditions was shown. The system can potentially be used for ranking cows by their efficiency in commercial facilities.

Dynamic cooling strategy based on individual animal response mitigated heat stress in dairy cows.

Technological progress enables individual cow's temperatures to be measured in real time, using a bolus sensor inserted into the rumen (reticulorumen). However, current cooling systems often work at a constant schedule based on the ambient temperature and not on monitoring the animal itself. This study hypothesized that tailoring the cooling management to the cow's thermal reaction can mitigate heat stress. We propose a dynamic cooling system based on in vivo temperature sensors (boluses). Thus, cooling can be activated as needed and is thus most efficacious. A total of 30 lactating cows were randomly assigned to one of two groups; the groups received two different evaporative cooling regimes. A control group received cooling sessions on a preset time-based schedule, the method commonly used in farms; and an experimental group, which received the sensor-based (SB) cooling regime. Sensor-based was changed weekly according to the cow's reaction, as reflected in the changes in body temperatures from the previous week, as measured by reticulorumen boluses. The two treatment groups of cows had similar milk yields (44.7 kg/d), but those in the experimental group had higher milk fat (3.65 vs 3.43%), higher milk protein (3.23 vs 3.13%), higher energy corrected milk (ECM, 42.84 vs 41.48 kg/d), higher fat corrected milk 4%; (42.76 vs 41.34 kg/d), and shorter heat stress duration (5.03 vs 9.46 h/day) comparing to the control. Dry matter intake was higher in the experimental group. Daily visits to the feed trough were less frequent, with each visit lasting longer. The sensor-based cooling regime may be an effective tool to detect and ease heat stress in high-producing dairy cows during transitional seasons when heat load can become severe in arid and semi-arid zones.

Effect of forage-to-concentrate ratio on production efficiency of low-efficient high-yielding lactating cows.

Feed is usually the costliest input in lactating cow's farms. Therefore, the developing of methods for a better adjustment of feed intake to cow's energetic needs in order to improve efficiency is desired. The aim of this study was to improve feed efficiency of low-efficient (LE) cows through a moderate increase in diet forage-to-concentrate ratio. We studied the effects of replacing 8.2% corn grains in a control low-fiber (LF) diet that contained 17.5% forage neutral detergent fiber (NDF) with 7.5% wheat straw +0.7% soybean meal for a high-fiber (HF) diet that contained 23.4% forage NDF. Based on efficiency data of individual cows from the Agricultural Research Organization's herd measured in our previous study, 15 pairs of pre-classified LE multiparous mid-lactating Israeli Holstein dairy cows were selected, each pair with similar performance, intake, and efficiency data; each member of a pair was then adapted for 2 weeks to one or the other dietary treatment. Traits examined during the 5 weeks of the experiment were DM intake (DMI), eating behavior, milk production, in vivo digestibility, and estimation of feed efficiency [energy-corrected milk (ECM)/DMI and energy balance]. Cows fed the HF diet showed slower eating rate, smaller visit and meal sizes, longer daily eating time, higher visit frequency, and longer meal duration, compared to those fed the LF diet. The DMI of cows fed the HF diet was 9.1% lower, their DM digestibility decreased from 65.7 to 62.2%, and their ECM yield was 7.0% lower than in cows fed the LF diet. Feed efficiency, measured as net energy captured/digestible energy intake, improved in the cows fed the HF vs. LF diet while feed efficiency measured as ECM/DMI remained similar. Our results thus show the potential of improving feed efficiency for milk production in LE cows by increasing the forage-to-concentrate ratio.

A machine vision system to detect and count laying hens in battery cages.

Manually counting hens in battery cages on large commercial poultry farms is a challenging task: time-consuming and often inaccurate. Therefore, the aim of this study was to develop a machine vision system that automatically counts the number of hens in battery cages. Automatically counting hens can help a regulatory agency or inspecting officer to estimate the number of living birds in a cage and, thus animal density, to ensure that they conform to government regulations or quality certification requirements. The test hen house was 87 m long, containing 37 battery cages stacked in 6-story high rows on both sides of the structure. Each cage housed 18 to 30 hens, for a total of approximately 11 000 laying hens. A feeder moves along the cages. A camera was installed on an arm connected to the feeder, which was specifically developed for this purpose. A wide-angle lens was used in order to frame an entire cage in the field of view. Detection and tracking algorithms were designed to detect hens in cages; the recorded videos were first processed using a convolutional neural network (CNN) object detection algorithm called Faster R-CNN, with an input of multi-angular view shifted images. After the initial detection, the hens' relative location along the feeder was tracked and saved using a tracking algorithm. Information was added with every additional frame, as the camera arm moved along the cages. The algorithm count was compared with that made by a human observer (the 'gold standard'). A validation dataset of about 2000 images achieved 89.6% accuracy at cage level, with a mean absolute error of 2.5 hens per cage. These results indicate that the model developed in this study is practicable for obtaining fairly good estimates of the number of laying hens in battery cages.

Dietary restriction improved feed efficiency of inefficient lactating cows.

The aim of this study was to reduce voluntary dry matter intake (DMI) to increase feeding efficiency of preclassified inefficient (INE) dairy cows through restricted feeding. We studied the effects of dietary restriction on eating behavior, milk and energy-corrected milk (ECM) production, in vivo digestibility, energy balance, and measures of feed efficiency [residual feed intake (RFI) and ECM/DMI]. Before the experiment, 12 pairs of cows were classified as INE. The 2 dietary treatments consisted of ad libitum feeding versus restricted feeding of the same total mixed ration containing 36.5% roughage. Inefficient cows fed the restricted total mixed ration had a shorter eating time and lower meal and visit frequency, but a similar rate of eating, meal size, and meal duration compared with INE cows fed ad libitum. Compared with the INE cows fed ad libitum, restricted INE cows had 12.8% lower intake, their dry matter and neutral detergent fiber digestibility remained similar, and their ECM yield was 5.3% lower. Feed efficiency, measured as RFI, ECM/DMI, and net energy retained divided by digestible energy intake, was improved in the restricted INE cows as compared with the ad libitum cows. Our results show that moderate DMI restriction has the potential to improve feed efficiency of preclassified INE cows.

Effect of lactation trimester and parity on eating behavior, milk production and efficiency traits of dairy cows.

There is absence knowledge about the effects of lactation trimester and parity on eating behavior, production and efficiency of dairy cows. Objective of this study was to identify and characterize in 340 dairy cows, the 20% high efficient (HE), 20% low efficient (LE) and 60% mid efficient (ME) cows according to their individual residual feed intake (RFI) values, within and between lactation trimesters and between 1st and 2nd parities. Efficiency effect within each lactation trimester, was exhibited in daily dry matter intake (DMI), eating rate and meal size, that were the highest in LE cows, moderate in the ME cows and lowest in the HE group. Daily eating time, meal frequency, yields of milk and energy-corrected milk (ECM) and BW were similar in the three efficiency groups within each trimester. The lower efficiency of the LE cows in each trimester attributes to their larger metabolic energy intake, heat production and energy losses. In subgroup of 52 multiparous cows examined along their 1st and 2nd trimesters, milk and ECM production, DMI, eating behavior and efficiency traits were similar with high Pearson's correlation (r=0.78 to 0.89) between trimesters. In another subgroup of 42 multiparous cows measured at their 2nd and 3rd trimesters, milk and ECM yield, DMI and eating time were reduced (P<0.01) at the 3rd trimester, but eating rate, meal frequency and meal size remained similar with high Pearson's correlation (r=0.74 to 0.88) between trimesters. In subgroup of 26 cows measured in 1st and 2nd parities, DMI, BW, milk and ECM yield, and ECM/DMI increased in the 2nd lactation, but eating behavior and RFI traits were similar in both parities. These findings encourage accurate prediction of DMI based on a model that includes eating behavior parameters, together with individual measurement of ECM production. This can be further used to identify HE cows in commercial herd, a step necessary for potential genetic selection program aimed to improve herd efficiency.

Eating behavior, milk production, rumination, and digestibility characteristics of high- and low-efficiency lactating cows fed a low-roughage diet.

This study aimed to identify individual characteristics differing among high-efficiency (HEf; upper 20%, n = 31), low-efficiency (LEf; lower 20%, n = 31), and mid-efficiency (MEf, 60% n = 93) lactating cows. Primiparous (37) and multiparous (118) high-producing milking cows at 30 to 180 d in milk were fed individually a low-roughage diet [31.6% of dry matter (DM)] for 4 wk. Daily average DM intake, rate of eating, visit duration, meal size, and daily rumination time were higher in LEf compared with HEf cows. On the other hand, HEf cows exhibited higher digestibility of DM, crude protein, and neutral detergent fiber than the LEf cows. Daily eating time was similar in the HEf and LEf groups and higher than that of the MEf cows. Visit and meal frequency, average visit and meal duration, daily lying time, and pedometer activity were similar in the HEf, LEf, and MEf groups. The HEf cows produced 1.75% more milk, but similar energy-corrected milk compared with the LEf cows. Milk fat and protein content were lower by 1.8 and 3.8%, respectively, in the HEf cows than in the LEf group. Body weight (BW) and BW gain were similar in the 3 efficiency groups. Diurnal distribution of DM intake showed 6 distinct major meals, each composed of 1.1 to 1.6 visits. Higher intake peaks (greater meal size) were found in the LEf cows compared with the HEf group. Daily DM intake was highly correlated (affected) with energy-corrected milk production (r = 0.61), BW (r = 0.4), eating rate (r = 0.57), and visit size (r = 0.54). Energy balance showed that the lower efficiency of the LEf cows was attributed to their excess heat production and energy loss.

Consistency of feed efficiency ranking and mechanisms associated with inter-animal variation among growing calves.

This study investigated the possible mechanisms for explaining interanimal variation in efficiency of feed utilization in intact male Holstein calves. Additionally, we examined whether the feed efficiency (FE) ranking of calves (n = 26) changed due to age and/or diet quality. Calves were evaluated during three periods (P1, P2, and P3) while fed a high-quality diet (calculated mobilizable energy [ME] of 11.8 MJ/kg DM) during P1 and P3, and a low-quality diet (calculated ME of 7.7 MJ/kg DM) during P2. The study periods were 84, 119, and 127 d, respectively. Initial ages of the calves in P1, P2, and P3 were 7, 11, and 15 mo, respectively, and initial body weight (BW) were 245, 367, and 458 kg, respectively. Individual dry matter intake (DMI), average daily gain (ADG), diet digestibility, and heat production (HP) were measured in all periods. The measured FE indexes were: residual feed intake (RFI), the gain-to-feed ratio (G:F), residual gain (RG), residual gain and intake (RIG), the ratio of HP-to-ME intake (HP/MEI), and residual heat production (RHP). For statistical analysis, animals' performance data in each period, were ranked by RFI, and categorized into high-, medium-, and low-RFI groups (H-RFI, M-RFI, and L-RFI). RFI was not correlated with in vivo digestibility, age, BW, BCS, or ADG in all three periods. The L-RFI group had lowest DMI, MEI, HP, retained energy (RE), and RE/ADG. Chemical analysis of the longissimus dorsi muscle shows that the L-RFI group had a higher percentage of protein and a lower percentage of fat compared to the H-RFI group. We suggested that the main mechanism separating L- from H-RFI calves is the protein-to-fat ratio in the deposited tissues. When efficiency was related to kg/day (DMI and ADG) and not to daily retained energy, the selected efficient L-RFI calves deposited more protein and less fat per daily gain than less efficient H-RFI calves. However, when the significant greater heat increment and maintenance energy requirement of protein compared to fat deposition in tissue were considered, we could not exclude the hypothesis that variation in efficiency is partly explained by efficient energy utilization. The ranking classification of calves to groups according to their RFI efficiency was independent of diet quality and age.

A flexible control system designed for lab-scale simulations and optimization of composting processes.

Understanding and optimization of composting processes can benefit from the use of controlled simulators of various scales. The Agricultural Research Organization Composting Simulator (ARO-CS) was recently built and it is flexibly automated by means of a programmable logic controller (PLC). Temperature, carbon dioxide, oxygen and airflow are monitored and controlled in seven 9-l reactors that are mounted into separate 80-l water baths. The PLC program includes three basic heating modes (pre-determined temperature profile, temperature-feedback ("self-heating"), and carbon dioxide-dependent temperature), three basic aeration modes (airflow dependence on temperature, carbon dioxide, or oxygen) and enables all possible combinations among them. This unique high flexibility provides a robust and valuable research tool to explore a wide range of research questions related to the science and engineering of composting. In this article the logic and flexibility of the control system is presented and demonstrated and its potential applications are discussed.

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.

Review: Milking robot utilization, a successful precision livestock farming evolution.

Automatic milking systems (AMS), one of the earliest precision livestock farming developments, have revolutionized dairy farming around the world. While robots control the milking process, there have also been numerous changes to how the whole farm system is managed. Milking is no longer performed in defined sessions; rather, the cow can now choose when to be milked in AMS, allowing milking to be distributed throughout a 24 h period. Despite this ability, there has been little attention given to milking robot utilization across 24 h. In order to formulate relevant research questions and improve farm AMS management there is a need to determine the current knowledge gaps regarding the distribution of robot utilization. Feed, animal and management factors and their interplay on levels of milking robot utilization across 24 h for both indoor and pasture-based systems are here reviewed. The impact of the timing, type and quantity of feed offered and their interaction with the distance of feed from the parlour; herd social dynamics, climate and various other management factors on robot utilization through 24 h are provided. This novel review draws together both the opportunities and challenges that exist for farm management to use these factors to improved system efficiency and those that exist for further research.

Feeding behavior improves prediction of dairy cow voluntary feed intake but cannot serve as the sole indicator.

Low-cost feeding-behavior sensors will soon be available for commercial use in dairy farms. The aim of this study was to develop a feed intake model for the individual dairy cow that includes feeding behavior. In a research farm, the individual cows' voluntary feed intake and feeding behavior were monitored at every meal. A feed intake model was developed based on data that exist in commercial modern farms: 'BW,' 'milk yield' and 'days in milking' parameters were applied in this study. At the individual cow level, eating velocity seemed to be correlated with feed intake (R 2=0.93 to 0.94). The eating velocity coefficient varied among individuals, ranging from 150 to 230 g/min per cow. The contribution of feeding behavior (0.28) to the dry matter intake (DMI) model was higher than the contribution of BW (0.20), similar to the contribution of fat-corrected milk (FCM)/BW (0.29) and not as large as the contribution of FCM (0.49). Incorporating feeding behavior into the DMI model improved its accuracy by 1.3 (38%) kg/cow per day. The model is ready to be implemented in commercial farms as soon as companies introduce low-cost feeding-behavior sensors on commercial level.

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.

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.