Bioacoustics features as a tool for early diagnosis of pneumonia in riverine buffalo (Bubalus bubalis) calves

Indu Devi. Livestock Production Management; Kuldeep Dudi. Animal Nutrition Division; Yajuvendra Singh. College of Veterinary Science and Animal Husbandry. Livestock Production Management; Surender Singh Lathwal. Livestock Research Centre. Livestock Production Management and In-Charge

Bioacoustics features as a tool for early diagnosis of pneumonia in riverine buffalo (Bubalus bubalis) calves

2021

Indu Devi(ICAR-Central Institute for Research on Cattle, Uttar Pradesh (India). Livestock Production Management) | Kuldeep Dudi(ICAR-National Dairy Research Institute, Haryana (India). Animal Nutrition Division) | Yajuvendra Singh(UP Pandit Deen Dayal Upadhyaya Pashu Chikitsa Vigyan Vishwavidyalaya Evam Go Anusandhan Sansthan, Uttar Pradesh (India). College of Veterinary Science and Animal Husbandry. Livestock Production Management) | Surender Singh Lathwal(Indian ICAR-National Dairy Research Institute, Haryana (India). Livestock Research Centre. Livestock Production Management and In-Charge)

The present study was conducted to identify specific acoustic features which could be used as possible indicators for early diagnosis of pneumonia in buffalo calves. In pneumonia, change in elasticity and resonance of vocal sound producing organs occur which eventually affects the vocal signature of calves. Twenty Murrah buffalo calves' voice was recorded during both healthy and pneumonia infected stage where pneumonia was confirmed by lung X-RAY radiography. From the recorded vocal sound, acoustic features viz. call duration (sec.), call interval (sec.), frequency (Hz), bandwidth (Hz) and peak amplitude (P) with their sub variants were extracted with the help of PRAAT 3.2.36 software. Out of these, call duration (sec.) (0.879+-0.29 v/s 0.689+-0.24), call interval (sec.) (0.288+-0.059 v/s 0.107+-0.047) and peak amplitude (P) (start (0.750+-0.118 v/s 0.435+-0.113), end (0.102+-0.045 v/s 0.508+-0.268) and maximum (0.938+-0.210 v/s 0.684+-0.480)) were found significantly (p LT 0.05) different between two groups. Rest acoustic features did not differ statistically between two groups. This study indicates that it is possible to discriminate pneumonia voice from normal/healthy voice by acoustic analysis and farmers can acquire an early warning of pneumonia infections in calves through this non-invasive method.

Show more [+]