The development and assessment of novel non-invasive methods of measuring sleep in dairy cows : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Animal Science at Massey University, Manawatū, New Zealand
Loading...
Date
2021
DOI
Open Access Location
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Massey University
Rights
The Author
Abstract
Sleep is important for animal health and welfare and there are many factors, for example, the environment, illness, or stress, that are likely to have an impact on cow sleep and consequently affect their welfare. Polysomnography (PSG) is considered the gold standard for precise identification of sleep stages. It consists of electrophysiological recordings of the brain activity, eye movements and muscle activity but is costly and difficult to use with cows on farm. Accordingly, the study of sleep in cows may be limited due to the impracticability of PSG. Alternative methods of assessing sleep have been developed for humans. Some such work has been conducted for cows, but this has yet to be validated with PSG.
The main aim of this thesis was to investigate alternative methods to PSG to accurately detect sleep stages in dairy cows. Specifically, I aimed to develop a detailed 5-stage scoring system for assessing sleep stages from the cow PSG, to investigate the suitability of using lying postures and heart rate (HR) measures to assess sleep stages and to develop a model to accurately predict sleep stages using non-invasive measures in dairy cows compared with PSG.
Two studies were conducted using 6 non-lactating dairy cows in an indoor housed environment in Scotland, and outdoors at pasture in New Zealand. PSG was recorded with each cow over a period of seven days. From these data a 5-stage sleep-scoring criteria with good reliability was developed which identified two stages of light sleep, two stages of deep sleep as well as awake and rumination stages.
Video was recorded during sleep recordings and the cow’s behaviour was analysed and compared with the scored sleep stages from the PSG. Some sleep stages were found to occur mainly in specific lying postures; however, overall, postures were not useful indicators of sleep stages. Heart rate (HR) and heart rate variability (HRV) were measured using a Polar HR monitor
ii
and data logging device. Differences in the HR and HRV measures were found between the sleep stages, and the patterns of these changes were similar between both Scottish and NZ cows. Finally, machine learning models were developed using supervised learning methods to predict sleep stage from the HR and HRV measures as well as the surface EMG data recorded during PSG. The models were able to learn to recognize and accurately predict sleep stages compared with the PSG scoring.
This research demonstrates that non-invasive alternatives can be used to identify sleep stages accurately in dairy cows compared with PSG. Further research is necessary with larger sample sizes and cows of various breeds and life stages; however, the success of the methods developed during this thesis demonstrates their suitability for the future measurement of sleep in cows and in the assessment of cow welfare.
Description
Onet published article in Appendix C was removed for copyright reasons, but may be accessed via its source: Hunter, L.B., O’Connor, C., Haskell, M.J., Langford, F.M., Webster, J.R., & Stafford, K.J. 2021, September. Lying posture does not accurately indicate sleep stage in dairy cows. Applied Animal Behaviour Science,
242, 105427. https://doi.org/10.1016/j.applanim.2021.105427
Keywords
Dairy cattle, Sleep, Measurement