Machine learning aids insights into cows’ sleep

Just like humans, our dairy cows need their sleep; but AgResearch scientists have had to get innovative with new technologies to better understand the quality of shut-eye the animals are getting.

Research into the sleep of cows is seen by researchers as one way of assessing their welfare, which is important to farmers and to respond to the growing expectations of global dairy consumers. Although little is known about the sleep needs of cows, scientists do know sleep is an essential physiological function for all animals and plays an important part in physical and mental health.

The challenge has been that measuring and distinguishing between the important stages of sleep in dairy cows is impractical with the animals housed in the usual farm environments, says AgResearch animal behaviour and welfare science team leader Dr Cheryl O’Connor.

“Between AgResearch and Scotland’s Rural College in Edinburgh, including joint PhD student Laura Hunter, we used sensor devices placed on the cows to take measurements during their sleep such as their neck muscle activity and heart rates, to compare with the gold standard EEG (electroencephalogram) for brain activity,” Dr O’Connor says.

“We took this muscle and heart rate data from six cows in both housed and pasture systems, and applied machine learning (a branch of Artificial Intelligence) models to make predictions about what the muscle and heart rate data means for the cows’ different sleep stages.”

“The result was that machine learning models were able to accurately predict sleep stages from the measures that were taken, and the accuracy was in a similar range to that for human computer models.”

Now that this method appears to be a valid way of measuring and predicting the sleep stages of cows at a small scale, researchers want to apply it to a much larger number of animals to validate the use of these methods.

“We think the insights we can get from this could potentially tell us more about overall animal welfare,” Dr O’Connor says.

“From that we may be able to build further on the research. We will be aiming to share what we learn with farmers and the wider industry, so they can potentially build that knowledge into what happens on farms to provide the best life we can for our cows.”

Publication

Machine learning prediction of sleep stages in dairy cows from heart rate and muscle activity measures

Sleep is important for cow health and shows promise as a tool for assessing welfare, but methods to accurately distinguish between important sleep stages are difficult and impractical to use with cattle in typical farm environments.

The objective of this study was to determine if data from more easily applied non-invasive devices assessing neck muscle activity and heart rate (HR) alone could be used to differentiate between sleep stages. We developed, trained, and compared two machine learning models using neural networks and random forest algorithms to predict sleep stages from 15 variables (features) of the muscle activity and HR data collected from 12 cows in two environments.

Using k-fold cross validation we compared the success of the models to the gold standard, Polysomnography (PSG). Overall, both models learned from the data and were able to accurately predict sleep stages from HR and muscle activity alone with classification accuracy in the range of similar human models.

Further research is required to validate the models with a larger sample size, but the proposed methodology appears to give an accurate representation of sleep stages in cattle and could consequentially enable future sleep research into conditions affecting cow sleep and welfare.

Read the full report here

Project Partner

Collaborator

Scotland's Rural College

Get in touch with our team

Contact us

Send an email to one of our team or check out our facilities located across Aotearoa New Zealand.

Send another enquiry

Something went wrong and the form could not be submitted. Please try again later.