Combination of Sensor Data and Health Monitoring for Early Detection of Subclinical Ketosis in Dairy Cows.

Valentin Sturm,Manfred Öhlschuster,Erika Gusterer,Dmitry Efrosinin,Marc Drillich,Michael Iwersen

doi:10.3390/s20051484

Abstract

Subclinical ketosis is a metabolic disease in early lactation. It contributes to economic losses because of reduced milk yield and may promote the development of secondary diseases. Thus, an early detection seems desirable as it enables the farmer to initiate countermeasures. To support early detection, we examine different types of data recordings and use them to build a flexible algorithm that predicts the occurence of subclinical ketosis. This approach shows promising results and can be seen as a step toward automatic health monitoring in farm animals.

Highlights

Subclinical ketosis (SCK) is a common metabolic disease of dairy cows in early lactation, characterised by an increased concentration of ketone bodies in the absence of clinical signs of disease [1]
We presented results from a study to identify indicators for subclinical ketosis in dairy cows around calving
We included a statistical comparison of different parameters, based on milk yield and components, animal movements within the barn, ambient temperature, and on visual observation

Summary

Introduction

Subclinical ketosis (SCK) is a common metabolic disease of dairy cows in early lactation, characterised by an increased concentration of ketone bodies in the absence of clinical signs of disease [1]. Analyzing the concentration of ß-hydroxybutyrate (BHB) in blood is the recommended reference test for detecting ketosis in dairy cows [2]. To detect SCK in dairy cows, various hand-held devices are commercially available, which were recently evaluated for use on farms [5,6]. The occurrence of SCK in dairy cows is associated with an increased risk of sequalae (e.g., clinical ketosis, displaced abomasum, metritis), decreased milk yield and impaired reproductive performance [3,7,8], affecting the economics of a dairy farm [9]. Recent studies showed that subclinical and clinical diseases are associated with distinct animal behaviours, e.g., rumination as well as with standing and lying times, respectively [11,12]

Objectives

Methods

Results

Conclusion