Livestock farming is currently undergoing a digital revolution and becoming increasingly data-driven. Yet, such data often reside in disconnected silos making them impossible to leverage their full potential to improve animal well-being. Here, we introduce a precision livestock farming approach, bringing together information streams from a variety of life domains of dairy cattle to study whether including more and diverse data sources improves the quality of predictions for eight diseases and whether using more complex prediction algorithms can, to some extent, compensate for less diverse data. Using three machine learning approaches of varying complexity (from logistic regression to gradient boosted trees) trained on data from 5,828 animals in 165 herds in Austria, we show that the prediction of lameness, acute and chronic mastitis, anestrus, ovarian cysts, metritis, ketosis (hyperketonemia), and periparturient hypocalcemia (milk fever) from routinely available data gives encouraging results. For example, we can predict lameness with high sensitivity and specificity (F1 = 0.74). An analysis of the importance of individual variables to prediction performance shows that disease in dairy cattle is a product of the complex interplay between a multitude of life domains, such as housing, nutrition, or climate, that including more and diverse data sources increases prediction performance, and that the reuse of existing data can create actionable information for preventive interventions. Our findings pave the way toward data-driven point-of-care interventions and demonstrate the added value of integrating all available data in the dairy industry to improve animal well-being and reduce disease risk.

This study aimed to develop a tool to detect mildly lame cows by combining already existing data from sensors, AMSs, and routinely recorded animal and farm data. For this purpose, ten dairy farms were visited every 30–42 days from January 2020 to May 2021. Locomotion scores (LCS, from one for nonlame to five for severely lame) and body condition scores (BCS) were assessed at each visit, resulting in a total of 594 recorded animals. A questionnaire about farm management and husbandry was completed for the inclusion of potential risk factors. A lameness incidence risk (LCS ≥ 2) was calculated and varied widely between farms with a range from 27.07 to 65.52%. Moreover, the impact of lameness on the derived sensor parameters was inspected and showed no significant impact of lameness on total rumination time. Behavioral patterns for eating, low activity, and medium activity differed significantly in lame cows compared to nonlame cows. Finally, random forest models for lameness detection were fit by including different combinations of influencing variables. The results of these models were compared according to accuracy, sensitivity, and specificity. The best performing model achieved an accuracy of 0.75 with a sensitivity of 0.72 and specificity of 0.78. These approaches with routinely available data and sensor data can deliver promising results for early lameness detection in dairy cattle. While experimental automated lameness detection systems have achieved improved predictive results, the benefit of this presented approach is that it uses results from existing, routinely recorded, and therefore widely available data.

The deployment of diverse data-generating technologies in livestock farming holds the promise of early disease detection and improved animal well-being. In this paper, we combine routinely collected dairy farm and herd data with weather and high frequency sensor data from 6 farms to predict new lameness events in various future periods, spanning from the following day to 3 weeks. A Random Forest classifier, using input features selected by the Boruta Algorithm, was used for the prediction task; effects of individual features were further assessed using partial dependence plots. We achieve precision scores of up to 93% when predicting lameness for the next 3 weeks and when using information from the last 3 weeks, combined with a balanced accuracy of 79%. Removing sensor data results have tendency to reduce the precision for predictions, especially when using information from the last one,2 or 3 weeks. Moving to a larger data set (without sensor data) of 44 farms keeps the similar balanced accuracy but reduces precision by more than 30%, revealing a substantial a trade-off in model quality between false positives (false lameness alerts) and false negatives (missed lameness events). Sensor data holds promise to further improve the precision of these models, but can be partially compensated by high resolution data from other systems, such as automated milking systems.

This study examines the relationship between efficiency, energy balance and related traits like milk yield, feed intake (DMI), body weight and body condition score (BCS). Data was derived in the project ‘Efficient Cow’ to develop efficiency traits for Austrian cattle breeding (6105 cows, 161 farms, one-year data collection). The following efficiency traits were considered: body weight efficiency as ratio between energy corrected milk (ECM) to metabolic body weight, feed efficiency (kg ECM per kg DMI) and energy efficiency expressed as ratio between energy in milk to energy intake. The higher the proportion of Red Holstein (RH) in Fleckvieh (FV), the more (efficiently) milk was produced, but also at the expenses of body fat reserves. The negative energy state of Holstein Friesian (HF) and the FV groups with highest RH proportion lasted approximately twice as long as of the least efficient Brown Swiss. All genotypes regained lost body tissue during whole lactation. The high yielding groups required a higher concentrate proportion in late lactation to regain body condition. In early lactation, high efficiency was accompanied by the loss of body weight and BCS. Body condition stagnated longer on the lowest level and was more conform to energy balance than body weight. In conclusion, high efficiency required an increasing partitioning of nutrients to milk yield inclusive mobilisation. Breeding for higher efficiency would exacerbate catabolic state including problems with health and fertility. This highlights the necessity of a broader definition of efficiency in cattle breeding involving parameters like BCS, health and fertility traits. HIGHLIGHTS

Modern dairy cattle breeding goals include several production and more and more functional traits. Estimated breeding values (EBV) that are combined in the total merit index usually come from single-trait models or from multivariate models for groups of traits. In most cases, a multivariate animal model based on phenotypic data for all traits is not feasible and approximate methods based on selection index theory are applied to derive the total merit index. Therefore, the objective of this study was to compare a full multitrait animal model with two approximate multitrait models and a selection index approach based on simulated data. Three production and two functional traits were simulated to mimic the national Austrian Brown Swiss population. The reference method for derivation of the total merit index was a multitrait evaluation based on all phenotypic data. Two of the approximate methods were variations of an approximate multitrait model that used either yield deviations or de-regressed breeding values. The final method was an adaptation of the selection index method that is used in routine evaluations in Austria and Germany. Three scenarios with respect to residual covariances were set up: residual covariances were equal to zero, or half of or equal to the genetic covariances. Results of both approximate multitrait models were very close to those of the reference method, with rank correlations of 1. Both methods were nearly unbiased. Rank correlations for the selection index method showed good results when residual covariances were zero but correlations with the reference method decreased when residual covariances were large. Furthermore, EBV were biased when residual covariances were high. We applied an approximate multitrait two-step procedure to yield deviations and de-regressed breeding values, which led to nearly unbiased results. De-regressed breeding values gave even slightly better results. Our results confirmed that ignoring residual covariances when a selection index approach is applied leads to remarkable bias. This could be relevant in terms of selection accuracy. Our findings suggest that the approximate multitrait approach applied to de-regressed breeding values can be used in routine genetic evaluation.