Development of an Early Detection System for Respiratory Diseases in Pigs

Abstract

Health issues are important in the management of group-housed livestock. In recent years, many studies have focused on reducing losses from respiratory diseases in livestock such as pigs, which can lead to death and reduced productivity. This study aimed to establish an early detection system to identify respiratory diseases in pigs using body-conducted sound (BCS). We aimed to elucidate the respiratory and heart sounds indicative of morbidity, such as pig respiratory disease. First, we developed a wireless recording system to record BCS, and features of respiratory sound were extracted from a signal obtained from the tip of the ear. An experiment was set up to record the BCS of pigs that had one type of respiratory disease – the porcine reproductive & respiratory syndrome. The acoustic analysis found significant differences before and after virus inoculation in zero crossing and mel-frequency cepstral coefficient acoustic features, which are the BCS features of pigs with a respiratory disease. Therefore, it is suggested that these acoustic features will help detect respiratory diseases early.