System specification and validation of a noseband pressure sensor for measurement of ruminating and eating behavior in stable-fed cows

Detailed specification of a novel scientific monitoring device for ruminating and eating activity in dairy cows.Evaluation of the measuring performance under field conditions in stable-fed cows.Convincing agreement between direct observations and automated measurements of chewing activity. Rumination and eating behavior are important indicators for assessing health and well-being in cattle. The objective of this study was to develop and validate a novel scientific monitoring device for automated measurement of ruminating and eating behavior in stable-fed cows to provide research with a measuring instrument for automated health and activity monitoring. The RumiWatch noseband sensor (Itin+Hoch GmbH, Liestal, Switzerland) incorporates a noseband pressure sensor, a data logger with online data analysis, and software. Automated measurements of behavioral parameters are based on generic algorithms without animal-specific learning data. Thereby, the system records and classifies the duration of chewing activities and enables users to quantify individual ruminating and eating jaw movements performed by the animal. During the course of the development, two releases of the system-specific software RumiWatch Converter (RWC) were created and taken into account for the validation study. The results generated by the two software versions, RWC V0.7.2.0 and RWC V0.7.3.2, were compared with direct behavioral observations. Direct observations of cow behavior were conducted on 14 Swiss dairy farms with an observation time of 1h per animal, resulting in a total sample of 60 dairy cows. Agreement of sensor measurement and direct observation was expressed as Spearman correlation coefficients (rs) for the pooled sample. For consolidated classification of sensor data (1-h resolution), correlations for rumination time were rs=0.91 (RWC V0.7.2.0) and rs=0.96 (RWC 0.7.3.2), and for eating time rs=0.86 (RWC 0.7.2.0) and rs=0.96 (RWC V0.7.3.2). Both software versions provide a high standard of validity and measuring performance for ruminating and eating behavior. The high to very high correlations between direct observation and sensor data demonstrate that the RumiWatch noseband sensor was successfully developed and validated as a scientific monitoring device for automated measurement of ruminating and eating activity in stable-fed dairy cows.