There is a critical need to develop and validate non-invasive animal-based indicators of affective states in livestock species, in order to integrate them into on-farm assessment protocols, potentially via the use of precision livestock farming (PLF) tools. One such promising approach is the use of vocal indicators. The acoustic structure of vocalizations and their functions were extensively studied in important livestock species, such as pigs, horses, poultry, and goats, yet cattle remain understudied in this context to date. Cows were shown to produce two types of vocalizations: low-frequency calls (LF), produced with the mouth closed, or partially closed, for close distance contacts, and open mouth emitted high-frequency calls (HF), produced for long-distance communication, with the latter considered to be largely associated with negative affective states. Moreover, cattle vocalizations were shown to contain information on individuality across a wide range of contexts, both negative and positive. Nowadays, dairy cows are facing a series of negative challenges and stressors in a typical production cycle, making vocalizations during negative affective states of special interest for research. One contribution of this study is providing the largest to date pre-processed (clean from noises) dataset of lactating adult multiparous dairy cows during negative affective states induced by visual isolation challenges. Here, we present two computational frameworks—deep learning based and explainable machine learning based, to classify high and low-frequency cattle calls and individual cow voice recognition. Our models in these two frameworks reached 87.2 and 89.4% accuracy for LF and HF classification, with 68.9 and 72.5% accuracy rates for the cow individual identification, respectively.

The present study aimed to employ machine learning algorithms based on sensor behavior data for (1) early-onset detection of digital dermatitis (DD) and (2) DD prediction in dairy cows. Our machine learning model, which was based on the Tree-Based Pipeline Optimization Tool (TPOT) automatic machine learning method, for DD detection on day 0 of the appearance of the clinical signs has reached an accuracy of 79% on the test set, while the model for the prediction of DD 2 days prior to the appearance of the first clinical signs, which was a combination of K-means and TPOT, has reached an accuracy of 64%. The proposed machine learning models have the potential to help achieve a real-time automated tool for monitoring and diagnosing DD in lactating dairy cows based on sensor data in conventional dairy barn environments. Our results suggest that alterations in behavioral patterns can be used as inputs in an early warning system for herd management in order to detect variances in the health and wellbeing of individual cows.

The overall objective of the current data report was to evaluate and test the feasibility of using infrared thermography (IRT) as a non-invasive method for measuring stress signs in lactating dairy cows during short negative challenges, such as visual isolation from herd-mates. The study was carried out at the Experimental Farm of the Research and Development Institute for Bovine Romania, on 20 Holstein-Friesian lactating multiparous dairy cows, between August and September 2022. Cows were housed in two identical tied stanchion barns (170/85 cm), and were isolated individually from the herd for 240 min post-morning milking. Our results shown significant (p ≤ 0.05) rises for both orbital and nasal IRT temperatures following the isolation challenge, suggesting that such approaches could represent adequate tools for assessing social stress in cattle. Overall, current results are in accordance with previous studies which validated both eye and nasal regions as IRT thermal windows for studying the effects of painful and negative contexts on stress response in farmed ruminants, while considering the stress-induced hyperthermia as an integral part of the physiological response to negative stimuli, as well as the current limitations that this tool faces.

The objective of the current pilot study was to evaluate the reproductive performance and health indicators in Dorper, White Dorper, and Tsigai breed ewes managed semi-intensively under European temperate conditions. A total of 544 ewe-year units were observed, with ewes (ranging from 1.5 to 8 years of age) managed under identical rearing conditions for a period of two consecutive production cycles (2012 through 2013 and 2013 through 2014). In general, significant (p≤0.001) genotype-related disparities were found in occurrence rates for all health parameters taken into study. Clinical mastitis incidence was significantly lower (p≤0.05) in Dorper (9.4%) and White Dorper (10.8%) breeds compared to that of Tsigai ewes (17.4%). Significant differences (p≤0.05) for lameness were found between Dorper and Tsigai breeds, with occurrence rates of 8.0% and 2.9%, respectively. Incidence for pneumonia and abortion was not influenced (p>0.05) by the ewes' genotype. Litter size was significantly lower (p≤0.05) in White Dorper breed than for Dorper and Tsigai ewes, of 1.21, 1.40, and 1.45, respectively. Conception rates and lambs survival were not affected (p>0.05) by genotype. Results suggest that South African Dorper and White Dorper sheep breeds have adapted well to the specific rearing conditions.

Aim of the current comparative study was to evaluate production outputs, reproduction efficiency and functional traits in dual-purpose Fleckvieh and Braunvieh cows, reared under temperate European conditions. A data-set from 414 Fleckvieh and 42 Braunvieh cows and 799 lactations was analysed. ID tag number, milk yield per milking session, number of steps/interval and milk conductivity were recorded and collected daily using AfiMilk 3.076 A-DU software (Afimilk Ltd., Kibbutz, Israel). Production and milk quality data were taken from the results of the official performance recordings and the reproductive outputs of cows were recorded by the research stations veterinarians. Comparisons between the two genotypes were carried out using the one way analysis of variance protocol, with categorical factor being considered the breed of cows. All the statistical inferences were carried out using Statistica software (StatSoft Inc., Tulsa, OK, USA). Fleckvieh cows significantly outperformed (p≤0.05) the Braunvieh herd, with average milk yields of 5,252.1±35.79 kg and 4,897.6±128.94 kg, respectively. Age at first calving was significantly (p≤0.01) influenced by the breed, with Fleckvieh heifers being more precocious (32.8±0.29 mo) compared to those of Braunvieh breed (35.7±0.84 mo). Reproduction efficiency as defined by the number of inseminations per gestation, calving interval, dystocia, days dry and days open, was not influenced by genotype (p>0.05). Incidences of sub-clinical mastitis, clinical mastitis, lameness and abortions were not influenced by the breed factor (p>0.05). Stayability of cows was significantly (p≤0.001) influenced by genotype, with Braunvieh cows having an average age at culling of 117.88±11.78 months compared to 90.88±2.89 months in Fleckvieh. Overall, results have shown that genotype significantly influenced milk yield, age at first calving and longevity.

Digital dermatitis (DD) is a common foot disease that can cause lameness, decreased milk production and fertility decline in cows. The prediction and early detection of DD can positively impact animal welfare and profitability of the dairy industry. This study applies deep learning-based computer vision techniques for early onset detection and prediction of DD using infrared thermography (IRT) data. We investigated the role of various inputs for these tasks, including thermal images of cow feet, statistical color features extracted from IRT images, and manually registered temperature values. Our models achieved performances of above 81% accuracy on DD detection on ‘day 0’ (first appearance of clinical signs), and above 70% accuracy prediction of DD two days prior to the first appearance of clinical signs. Moreover, current findings indicate that the use of IRT images in conjunction with AI based predictors show real potential for developing future real-time automated tools to monitoring DD in dairy cows.