Rapid detection of dairy cow mastitis is important so corrective action can be taken as soon as possible. Automatically collected sensor data used to monitor the performance and the health state of the cow could be useful for rapid detection of mastitis while reducing the labor needs for monitoring. The state of the art in combining sensor data to predict clinical mastitis still does not perform well enough to be applied in practice. Our objective was to combine a multivariate dynamic linear model (DLM) with a naïve Bayesian classifier (NBC) in a novel method using sensor and nonsensor data to detect clinical cases of mastitis. We also evaluated reductions in the number of sensors for detecting mastitis. With the DLM, we co-modeled 7 sources of sensor data (milk yield, fat, protein, lactose, conductivity, blood, body weight) collected at each milking for individual cows to produce one-step-ahead forecasts for each sensor. The observations were subsequently categorized according to the errors of the forecasted values and the estimated forecast variance. The categorized sensor data were combined with other data pertaining to the cow (week in milk, parity, mastitis history, somatic cell count category, and season) using Bayes' theorem, which produced a combined probability of the cow having clinical mastitis. If this probability was above a set threshold, the cow was classified as mastitis positive. To illustrate the performance of our method, we used sensor data from 1,003,207 milkings from the University of Florida Dairy Unit collected from 2008 to 2014. Of these, 2,907 milkings were associated with recorded cases of clinical mastitis. Using the DLM/NBC method, we reached an area under the receiver operating characteristic curve of 0.89, with a specificity of 0.81 when the sensitivity was set at 0.80. Specificities with omissions of sensor data ranged from 0.58 to 0.81. These results are comparable to other studies, but differences in data quality, definitions of clinical mastitis, and time windows make comparisons across studies difficult. We found the DLM/NBC method to be a flexible method for combining multiple sensor and nonsensor data sources to predict clinical mastitis and accommodate missing observations. Further research is needed before practical implementation is possible. In particular, the performance of our method needs to be improved in the first 2 wk of lactation. The DLM method produces forecasts that are based on continuously estimated multivariate normal distributions, which makes forecasts and forecast errors easy to interpret, and new sensors can easily be added.

The preferred habitat of a given bacterium can provide a hint of which types of enzymes of potential industrial interest it might produce. These might include enzymes that are stable and active at very high or very low temperatures. Being able to accurately predict this based on a genomic sequence, would thus allow for an efficient and targeted search for production organisms, reducing the need for culturing experiments.

Accurate forecasting of dairy cow milk yield is useful to dairy farmers, both in relation to financial planning and for detection of deviating yield patterns, which can be an indicator of mastitis and other diseases. In this study we developed a dynamic linear model (DLM) designed to forecast milk yields of individual cows per milking, as they are milked in milking robots. The DLM implements a Wood's function to account for the expected total daily milk yield. It further implements a second-degree polynomial function to account for the effect of the time intervals between milkings on the proportion of the expected total daily milk yield. By combining these 2 functions in a dynamic framework, the DLM was able to continuously forecast the amount of milk to be produced in a given milking. Data from 169,774 milkings on 5 different farms in 2 different countries were used in this study. A separate farm-specific implementation of the DLM was made for each of the 5 farms. To determine which factors would influence the forecast accuracy, the standardized forecast errors of the DLM were described with a linear mixed effects model (lme). This lme included lactation stage (early, middle, or late), somatic cell count (SCC) level (nonelevated or elevated), and whether or not the proper farm-specific version of the DLM was used. The standardized forecast errors of the DLM were only affected by SCC level and interactions between SCC level and lactation stage. Therefore, we concluded that the implementation of Wood's function combined with a second-degree polynomial is useful for dynamic modeling of milk yield in milking robots, and that this model has potential to be used as part of a mastitis detection system.

Tail biting in pigs is an animal welfare problem, and tail biting should be prevented from developing into tail damage. One strategy could be to predict events of tail biting so that the farmer can make timely interventions in specific pens. In the current investigation, sensor data on water usage (water flow and activation frequency) and pen temperature (above solid and slatted floor) were included in the development of a prediction algorithm for tail biting. Steps in the development included modelling of data sources with dynamic linear models, optimisation and training of artificial neural networks and combining predictions of the single data sources with a Bayesian ensemble strategy. Lastly, the Bayesian ensemble combination was tested on a separate batch of finisher pigs in a real-life setting. The final prediction algorithm had an AUC > 0.80, and thus it does seem possible to predict events of tail biting from already available sensor data. However, around 30% of the no-event days were false alarms, and more event-specific predictors are needed. Thus, it was suggested that farmers could use the alarms to point out pens that need greater attention.

Reticuloruminal pH has been linked to subclinical disease in dairy cattle, leading to considerable interest in identifying pH observations below a given threshold. The relatively recent availability of continuously monitored data from pH boluses gives new opportunities for characterizing the normal patterns of pH over time and distinguishing these from abnormal patterns using more sensitive and specific methods than simple thresholds. We fitted a series of statistical models to continuously monitored data from 93 animals on 13 farms to characterize normal variation within and between animals. We used a subset of the data to relate deviations from the normal pattern to the productivity of 24 dairy cows from a single herd. Our findings show substantial variation in pH characteristics between animals, although animals within the same farm tended to show more consistent patterns. There was strong evidence for a predictable diurnal variation in all animals, and up to 70% of the observed variation in pH could be explained using a simple statistical model. For the 24 animals with available production information, there was also a strong association between productivity (as measured by both milk yield and dry matter intake) and deviations from the expected diurnal pattern of pH 2 d before the productivity observation. In contrast, there was no association between productivity and the occurrence of observations below a threshold pH. We conclude that statistical models can be used to account for a substantial proportion of the observed variability in pH and that future work with continuously monitored pH data should focus on deviations from a predictable pattern rather than the frequency of observations below an arbitrary pH threshold.