Graduate Student Literature Review: The use of integrated sensor data for the detection of hyperketonemia in pasture-based dairy systems during the transition period.

This review evaluates research regarding the use of sensors to predict and manage hyperketonemia (HYK) in dairy cows during the transition period, with a focus on pasture-based systems. By doing so, we assessed the accuracy of HYK-detection models, noting that no studies thus far have produced models with sufficient accuracy for practical use. Sensors have been validated for their use in dairy farming, proving they produce reliable and useful information. Research is beginning to focus on the analysis of multiple sensors together as a sensor system, discovering the potential for these technologies to be a valuable aid in decision making and farm management. Of the studies that use sensors to predict and manage disease in dairy cows, few studies use data integration (the process of combining data from multiple sensors which in turn improves model accuracy), highlighting a gap in the literature. Recently published research has focused on the detection of mastitis and lameness in pasture-based systems, with less focus toward the detection of metabolic disease. This is reflected in the lack of studies that report the prevalence of metabolic diseases, such as HYK, in pasture-based systems, especially in Australia and New Zealand. It is suggested that further research focuses on (1) determining the prevalence and effect of HYK in pasture-based systems; (2) exploring the use of sensors for HYK detection in pasture-based systems; (3) improving model accuracy with data integration; and (4) confirming the economic benefit of sensors to justify the cost of investing in sensor systems.