Measuring dairy cow welfare with real-time sensor-based data and farm records: a concept study

Welfare assessment of dairy cows by in-person farm visits provides only a snapshot of welfare and is time-consuming and costly. Possible solutions to reduce the need for in-person assessments would be to exploit sensor data and other routinely collected on-farm records. The aim of this study was to develop an algorithm to classify dairy cow welfare based on sensors (accelerometer and/or milk meter) and farm records (e.g. days in milk, lactation number). In total, 318 cows from six commercial farms located in Finland, Italy and Spain (two farms each) were enrolled for a pilot study lasting 135 days. During this time, cows were routinely scored using 14 animal-based measures of good feeding, health and housing based on the Welfare Quality® (WQ®) protocol. WQ® measures were evaluated daily or approximately every 45 days, using disease treatments from farm records and on-farm visits, respectively. WQ® measures were supplemented with daily temperature-humidity index to account for heat stress. The severity and duration of each welfare measure were evaluated, and the final welfare index was obtained by summing up the values for each cow on each pilot study day, and stratifying the result into three classes: good, moderate and poor welfare. For model building, a machine-learning (ML) algorithm based on gradient-boosted trees (XGBoost) was applied. Two model versions were tested: (1) a global model tested on unseen herd, and (2) a herd-specific model tested on unseen part of the data from the same herd. The version (1) served as an example on the model performance on a herd not previsited by the evaluator, while version (2) resembled a custom-made solution requiring in-person welfare evaluation for model training. Our results indicated that the global model had a low performance with average sensitivity and specificity of 0.44 and 0.68, respectively. For the herd-specific version, the model performance was higher reaching an average of 0.64 sensitivity and 0.80 specificity. The highest classification performance was obtained for cows in poor welfare, followed by cows in good and moderate welfare (balanced accuracy of 0.77, 0.71 and 0.68, respectively). Since the global model had low classification accuracy, the use of the developed model as a stand-alone system based solely on sensor data is infeasible, and a combination of in-person and sensor-based welfare evaluation would be preferable for a reliable welfare assessment. ML-based solutions, even with fair discriminative abilities, have the potential to enhance dairy welfare monitoring.

Global and regional longitudinal strain assessed by two-dimensional speckle tracking echocardiography identifies early myocardial dysfunction and transmural extent of myocardial scar in patients with acute ST elevation myocardial infarction and relatively preserved LV function

AIMS

Global and regional longitudinal strain (GLS-RLS) assessed by two-dimensional speckle tracking echocardiography (2D-STE) are considered reliable indexes of left-ventricular (LV) function and myocardial viability in chronic ischaemic patients when compared with delayed-enhanced cardiac magnetic resonance (DE-CMR). In the present study, we tested whether GLS and RLS could also identify early myocardial dysfunction and transmural extent of myocardial scar in patients with acute ST elevation myocardial infarction (STEMI) and relatively preserved LV function.

METHODS AND RESULTS

Twenty STEMI patients with LVEF ≥40%, treated with PPCI within 6 h from symptoms onset, underwent DE-CMR and 2D-echocardiography for 2D-STE analysis 6 ± 2 days after STEMI. Wall motion score index (WMSI) and LV ejection fraction (LVEF) were calculated by both methods. Infarct size and transmural extent of necrosis were assessed by CMR. GLS and RLS were obtained by 2D-STE. Mean GLS of the study population was -14 ± 3.3, showing a significant correlation with both LVEF and WMSI, by CMR (r = -0.86, P = 0.001, and r = 0.80, P = 0.001, respectively) and time-to-PCI (r = 0.66, P = 0.038). A weaker correlation was found between GLS and LVEF and WMSI assessed by 2D-echo (r = -0.65, P = 0.001, and r = 0.53, P = 0.013, respectively). RLS was significantly lower in DE-segments when compared with normal myocardium (P < 0.0001). A cut-off value of RLS of -12.3% by receiver-operating characteristic (ROC) curves identified DE-segments (sensitivity 82%, specificity 78%), whereas a cut-off value of -11.5% identified transmural extent of DE (sensitivity 75%, specificity 78%).

CONCLUSION

Our findings indicate that RLS and GLS evaluation provides an accurate assessment of global myocardial function and of the presence of segments with transmural extent of necrosis, with several potential clinical implications.

The effect of transport by road and sea on physiology, immunity and behaviour of beef cattle

The objective of the study was to investigate the physiological, haematological and immunological responses of weanling heifers transported from Ireland to a feedlot in Spain, and of weanling bulls transported from Ireland to a feedlot in Italy. Physiological variables (including interferon-γ production, cortisol, protein, urea, white blood cell numbers and differentials, and acute phase proteins (haptoglobin and fibrinogen) were used to evaluate the welfare status of animals, before, during and after the respective transport journeys. Age-matched control animals were blood sampled for the same measurements at times corresponding to the transported animals that were retained in Ireland. Heifers transported to Spain lost 7.6% of their initial live weight during the sea crossing to France. However, by the time of their arrival in Spain they had regained 3.3% of their initial live weight and had fully recovered to their pre-transport live weight values within 6 days of arriving in Spain. Weanling bulls lost 7.0% of their live weight during the sea crossing from Ireland to France. The live weight loss in control animals ranged from 1% to 2% during the same period. The percentage of time that bulls spent lying was 63.5% for the sea journey and 35.4% for the journey from the French lairage to the Italian feedlot. The average daily gain (kg) of transported animals was greater (P ≤ 0.05) than control animals from day 11 to 38 (Spain) and day 11 to 40 (Italy), respectively. While transient changes in physiological, haematological and immunological variables were found in the transported and control animals relative to baseline levels, the values were within the normal physiological range for the age and weight of animals involved. Physiological measurements made after the road and sea journeys indicated that the 24h rest in the lairage, with hay and water freely available, allowed animals to recover substantially.

A critical review of fear tests used on cattle, pigs, sheep, poultry and horses

Fear is arguably the most commonly investigated emotion in domestic animals. In the current review we attempt to establish the level of repeatability and validity found for fear tests used on cattle, pigs, sheep and goats, poultry and horses. We focus the review on the three most common types of fear tests: the arena test (open field), the novel object test, and the restraint test. For some tests, e.g. tonic immobility in poultry, there is a good and broad literature on factors that affect the outcome of the test, the validity of the test and its age dependency. However, there are comparatively few of these well defined and validated tests and what is especially missing for most tests is information on the robustness, i.e., what aspects can be changed without affecting the validity of the tests. The relative absence of standardized tests hampers the development of applied ethology as a science.

The provision of drinking water to veal calves for welfare purposes

Growth performance, behavior, physiology, forestomach development, abomasal lesions, and meat quality of veal calves fed a milk-replacer diet (No Water) were compared to those obtained from calves fed the same diet and provided with increasing amounts of drinking water (Water). Two groups of 69 Polish Friesian calves, balanced according to initial BW, were assigned to two water treatments in a 3 x 2 x 2 factorial arrangement that provided solid feed in addition to the milk-replacer diet (No solid feed, 250 g x calf(-1) x d(-1) of wheat straw or the same amount of beet pulp), and the adoption of two housing systems (individual stall vs group pen). The fattening trial lasted 160 d, and calves received drinking water starting from the 2nd wk of the study. The amount of drinking water was progressively increased from 3 to 8 L x cal(-1) x d(-1). Although not dehydrated, as shown by hematocrit and Na, K, and total protein hemoconcentration, calves consumed almost all the offered amount of water throughout the fattening period. Therefore, the water provided by the milk replacer alone, which ranged between 6 to 16 L calf d(-1), was not sufficient to satisfy the need of the animal. Drinking water did not affect the calves' growth performance but it reduced nonnutritive oral behavior throughout the fattening period. Based on these results, drinking water did not cover a shortage in the calves' water requirement but it played a role in environmental enrichment. Health status was similar between treatments, although water provision reduced the episodes of feed refusal. The measurement of chronic stress by ACTH challenge showed that the administration of drinking water would be advisable when calves are fed with small amounts of solid feed for well-being. Feces consistency and animal cleanliness were not affected by drinking water. At slaughter, forestomach development was similar between treatments, and drinking water did not affect the number of calves showing rumen hair-balls and abomasal lesions. No differences in color and other meat quality traits were observed between Water and No Water calves. Despite the lack of direct effects on productive traits, when water was available, the calves drank it, and positive effects were noticed on their nonnutritive oral behaviors and chronic stress indicators.

The provision of solid feeds to veal calves: II. Behavior, physiology, and abomasal damage

The aim of this study was to evaluate the effects of the addition of two roughage sources (wheat straw and beet pulp) to the milk replacer diet of veal calves, in order to reduce stress and improve animal welfare. We allocated 138 Polish Friesian male calves to three different feeding plans: a milk replacer diet (Control), 250 g/d of wheat straw in addition to the milk replacer, or 250 g/d of dried beet pulp in addition to the milk replacer. Within each feeding treatment, 16 calves were individually housed and 30 were kept in group pens (five calves/pen). Several behavioral, physiological, and health welfare indicators were monitored throughout the fattening period, which lasted for 160 d. Abnormal oral behavior around the meals was higher in Control calves (P < 0.01), while its lowest level was observed in straw-fed calves. At the beginning of the trial, chewing was higher in calves receiving solid feeds (P < 0.001), but the difference from the Control gradually decreased and disappeared at wk 13 for calves fed beet pulp and at wk 17 for those fed wheat straw. At the end of the fattening period, no differences among treatments were found in the frequency of chewing. Regardless of the diet, self-grooming decreased with age and no relationship was observed between this behavior and the presence of rumen hairballs. Cross-sucking was performed with low frequencies (from 4.70% at wk 2 to 1.05% at wk 23 around the meals, and even lower far from the meals) and was not affected by the provision of roughage. The time in contact with the bucket during the whole day was higher in Controls, whereas calves fed wheat straw maintained a lower level of this activity until the end of the trial (P < 0.01). The calves fed wheat straw spent more time in contact with the feed trough (P < 0.001) than those fed beet pulp and Control calves. No differences were found in cortisol curves due to the feeding treatment. In calves fed beet pulp, most hematological measures statistically differed from the other treatments, possibly in response to the higher iron intake and(or) to the higher hemoconcentration, probably due to the administration of beet pulp as dried feed. The incidence of abomasal ulcers and erosions was increased by the provision of the solid feeds, particularly by a structured fiber source such as straw. A roughage source able to satisfy calves' behavioral needs and to improve digestive processes without damaging the digestive apparatus still has to be identified.

The provision of solid feeds to veal calves: I. Growth performance, forestomach development, and carcass and meat quality

Growth performance, forestomach development, and carcass and meat quality of veal calves fed a milk replacer diet (Control) were compared to those obtained from calves fed the same liquid diet plus 250 g x calf(-1) x d(-1) of dried beet pulp or wheat straw. Three groups of 46 Polish Friesian calves, balanced according to initial BW, were assigned to the three dietary treatments in a fattening trial, which lasted 160 d. The provision of either solid feed did not affect the milk replacer intake. However, calves' ADG was increased (P < 0.01) only by feeding the beet pulp diet. The administration of both solid feeds improved calves' health status; calves fed solid feeds required fewer iron treatments for low hemoglobin and needed less medical treatments for respiratory or gastrointestinal diseases. In comparison to the Control calves, the provision of wheat straw and beet pulp increased iron intake throughout the fattening period by 41 and 130%, respectively. However, only calves fed beet pulp showed higher levels of hemoglobin and plasma iron concentrations (P < 0.05), whereas the same blood parameters were similar between Control calves and those fed wheat straw. At slaughter, both solid feeds led to empty forestomach weights heavier than those of Controls without reducing dressing percentage. The reticulorumen was heaviest in calves fed beet pulp, whereas wheat straw promoted omasal development. The administration of beet pulp resulted in a better carcass conformation than did the Control diet or wheat straw, but it had a detrimental effect on carcass color, which was graded as the darkest (P < 0.001). Consistent with this result, meat color of calves fed beet pulp was darker than that of Control calves and those fed wheat straw, because of the higher hematin concentration measured at the muscle level. No differences in carcass and meat color were observed between Control calves and calves fed wheat straw. The administration of solid feeds for welfare purposes does not always prevent the production of veal meat fulfilling the color standards required by the market. There is not a straight-forward relationship between a solid feed's iron content and the "redness" of veal meat, which should be related to the capability of the calves to use the iron provided by the roughage.

Cilia-Associated Respiratory (CAR) bacillus infection in veal calves and adult cattle

Cilia-Associated Respiratory (CAR) bacillus is a filamentous bacterium that colonizes the ciliated epithelium of the respiratory tract of many animal species and that has been associated with chronic inflammatory lesions in naturally and experimentally infected rats, mice and rabbits. In the present study, the prevalence of CAR bacillus infection and histological lesions of the trachea in veal calf and adult cattle were investigated. Forty five healthy veal calves and 45 adult cattle, raised in 18 different herds were selected at slaughter. From each animal, a tracheal sample was processed for histology, stain-ed with the Warthin-Starry method to evaluate the presence of CAR bacillus, and with haematoxylin and eosin to evaluate the presence of inflammatory lesions. CAR bacillus was identified in 17 veal calves (37.7%) and in 7 adult cattle (15.5%). Inflammatory lesions were found in 42 veal calves (93.3%) and in 41 adult cattle (91.1%). Statistical analysis pointed out a significant correlation between the presence and number of CAR bacilli and the presence and number of lymphoid follicles (P = 0.0071) and the presence and severity of neutrophilic infiltrates (P = 0.0428). These results indicate that CAR bacillus infection is common in cattle and is correlated with tracheal inflammatory lesions.

Heart rate and behavioural responses of crib-biting horses to two acute stressors

The heart rate and behaviour of 14 adult saddle horses, eight crib-biters and six normal controls, were investigated. Initially, the relationship between crib-biting and heart rate was investigated while the horses were undisturbed. The horses were tested when restrained with a lip twitch, and assessed when they were exposed suddenly to the rapid inflation of a balloon. The heart rate of the crib-biters during crib-biting was lower than during other behaviours. The crib-biters had a higher overall mean heart rate (P<0.05) suggesting that they may have had a higher basal sympathetic activity. After the application of the twitch, all the horses had a transient increase in heart rate which returned to basal values more rapidly in the crib-biters. The crib-biters were less reactive to the lip twitch, five of the six investigated remaining calm, and after the release of the twitch, they spent more time nibbling (P<0.05) than the control horses. The crib-biters reacted more strongly to the inflation of the balloon (three of the six reacted), and after it had been inflated they spent more time walking in the box.

Effects of blood sampling procedures, grouping and adrenal stimulation on stress responses in the growing pig

Changes in the blood and the behaviour of 14 growing pigs from 4 different litters were evaluated under different experimental conditions of blood sampling, grouping and adrenal stimulation. The results showed that the different techniques of blood sampling influenced lactic dehydrogenase (LDH) and creatine kinase (CK) activities. Cortisol, proteins and CK levels were negatively correlated with social hierarchy after regrouping. Cortisol was also correlated with total activity levels. Adrenal stimulation by adrenocorticotropic hormone (ACTH) administration caused a sharp increase in plasma cortisol levels. However, plasma glucose, plasma proteins and total leukocyte counts were not affected by the ACTH treatment.