RESUMEN
The aim of the present study was to develop a model to identify posture and behavior from data collected by a triaxial accelerometer located on the left flank of dairy cows and evaluate its accuracy and precision. Twelve Italian Red-and-White lactating cows were equipped with an accelerometer and observed on average for 136 ± 29 min per cow by two trained operators as a reference. The acceleration data were grouped in time windows of 8 s overlapping by 33.0%, for a total of 35,133 rows. For each row, 32 different features were extracted and used by machine learning algorithms for the classification of posture and behavior. To build up a predictive model, the dataset was split in training and testing datasets, characterized by 75.0 and 25.0% of the observations, respectively. Four algorithms were tested: Random Forest, K Nearest Neighbors, Extreme Boosting Algorithm (XGB), and Support Vector Machine. The XGB model showed the best accuracy (0.99) and Cohen's kappa (0.99) in predicting posture, whereas the Random Forest model had the highest overall accuracy in predicting behaviors (0.76), showing a balanced accuracy from 0.96 for resting to 0.77 for moving. Overall, very accurate detection of the posture and resting behavior were achieved.
RESUMEN
This study aimed at assessing whether a reduction in stocking density (SKD) would mitigate lameness and positively affect the performance and health of Charolais bulls in an Italian commercial farm. Bulls were distributed in groups of 10 or 8 animals/pen for high (HD) or low density (LD) corresponding to an individual space of 3.5 or 4.7 m2, respectively. Bulls were fitted with collars that measured rumination time and activity. Three 8-h observational sessions were conducted to record behaviors. Data about health conditions were collected daily. No differences were found in the animals' performance. However, performance results might have been impaired by the culling rate experienced during the trial, which prevented from keeping a consistent SKD. Behaviors did not differ between groups, except for rumination time, which was higher for LD bulls during the third observation (p < 0.05). However, rumination time, recorded by collars, did not vary among treatments. There were no differences in the percentage of sick or lame bulls, but the percentage of animals treated repeatedly due to relapse was higher for the HD group (p < 0.05). It was concluded that a larger space allowance could improve the health of bulls kept on fully slatted floors.
RESUMEN
This study aimed to evaluate the effect of decreasing dietary crude protein (CP) on the performance of finishing Charolais bulls in the Italian rearing system. Animals were fed two diets, differing only in the CP level (low protein (LP), 13.5% CP versus control (CON), 15.0% CP). Dry matter (DM) intake (DMI) and animals' weights were recorded to obtain average daily gain (ADG) and feed conversion rate (FCR). Feed and fecal samples were collected to evaluate digestibility of diet components. Daily cost of the ration (DRC), feed cost per kg of daily weight gain (CDG) and daily gross margin (DGM) were calculated to analyze the possible benefits of decreasing the protein level. Meat quality analyses were also conducted. Higher DMI (10.6 versus 10 kg/d; p < 0.05) and ADG (1.47 versus 1.36 kg/d; p < 0.05) were observed for CON. No differences in FCR or digestibility were found. Even if the DRC was lower (p < 0.05) for the LP diet (2.26 versus 1.97 ; CON versus LP), no difference was reported for CDG and DGM. Meat lightness and redness were significantly lower and higher in the LP, respectively. To conclude, the CP requirement in these rearing conditions appeared to be higher than 13.5%.