This research communication of technical interest addresses the goal to validate an algorithm to monitor natural occurrence of locomotor behaviours in dairy Mediterranean buffalo based on the output of a 3-dimensional accelerometer (RumiWatch®, pedometer). Several characteristics of the locomotor behaviour were detected with a very high (up-right, lying and standing time) or high degree of correlation (walking time and number of strides) and a low mean difference with the video recording. The proportion of correctly detected events exceeded 99% for the following variables: stand up and lie down events, as well as number of lying, standing or walking bouts. The mean relative measurement error was less than 10% for the variables: lying, standing, up-right times and number of strides as compared with gold standard. This new algorithm may represent the base for a future early and real-time disease warning system aiming to gain higher health standard in these ruminants.
Validation of a pedometer algorithm as a tool for evaluation of locomotor behaviour in dairy Mediterranean buffalo / D'Andrea, Luigi; Guccione, Jacopo; Alsaaod, Maher; Deiss, Ramona; DI LORIA, Antonio; Steiner, Adrian; Ciaramella, Paolo. - In: JOURNAL OF DAIRY RESEARCH. - ISSN 1469-7629. - 84:4(2017), pp. 391-394. [10.1017/S0022029917000668]
Validation of a pedometer algorithm as a tool for evaluation of locomotor behaviour in dairy Mediterranean buffalo.
D'andrea Luigi;Guccione Jacopo
;Di Loria Antonio;Ciaramella Paolo
2017
Abstract
This research communication of technical interest addresses the goal to validate an algorithm to monitor natural occurrence of locomotor behaviours in dairy Mediterranean buffalo based on the output of a 3-dimensional accelerometer (RumiWatch®, pedometer). Several characteristics of the locomotor behaviour were detected with a very high (up-right, lying and standing time) or high degree of correlation (walking time and number of strides) and a low mean difference with the video recording. The proportion of correctly detected events exceeded 99% for the following variables: stand up and lie down events, as well as number of lying, standing or walking bouts. The mean relative measurement error was less than 10% for the variables: lying, standing, up-right times and number of strides as compared with gold standard. This new algorithm may represent the base for a future early and real-time disease warning system aiming to gain higher health standard in these ruminants.File | Dimensione | Formato | |
---|---|---|---|
Rumiwatch Podometro bufala.pdf
solo utenti autorizzati
Tipologia:
Documento in Post-print
Licenza:
Accesso privato/ristretto
Dimensione
88.99 kB
Formato
Adobe PDF
|
88.99 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.