Combined analysis of grazing behaviour and movement: a step towards automated lameness detection in cattle

Context and issues

Lameness is one of the most problematic conditions in dairy cattle farming, with a strong impact on animal welfare and significant economic losses for farmers ^1,2. Farmers traditionally detect lame cows by watching them move, especially towards the milking parlour. Unfortunately, many lame cows go undetected or are detected late by farmers, which greatly compromises their chances of recovery ³. 3 Automated lameness detection using monitoring tools could improve detection. Accelerometers are already widely used on farms to help detect oestrus, but their performance is currently insufficient to detect lame cows, especially when they are on pasture ^4,5. The combination of information collected by an accelerometer and a GPS on a collar has shown behavioural and movement changes in lame cows and therefore to automate the detection of lameness on pasture.

Results

This study confirms that several behavioural and movement parameters are modified in lame cows at pasture. Non-lame cows spent 4.5 and 1.6 times more time ingesting grass than severely and moderately lame cows, respectively, with longer ingestion periods. Severely lame cows spent almost twice as much time resting as non-lame cows, particularly in the lying down position, with both more rest periods and longer rest periods. Lame cows also laid down twice as soon after entering the plot as non-lame cows or with a slightly modified gait. Finally, non-lame cows walked 1.5 times more distance on pasture than lame cows and scattered 1.6 times more than severely lame cows.

Perspectives

This study provides encouraging results on the use of a combined accelerometer and GPS tool to automate lameness detection in grazing dairy cows. However, these tools should be combined with other tools to better detect moderately lame cows, including those without access to pasture. The frequency of readings and battery life will also need to be optimised to make it a tool that farmers can use on a daily basis.

Valorisation

Riaboff, L., Relun, A., Petiot, C.E., Feuilloy, M., Couvreur, S., Madouasse, A., 2021. Identification of discriminating behavioural and movement variables in lameness scores of dairy cows at pasture from accelerometer and GPS sensors using a Partial Least Squares Discriminant Analysis. Prev. Vet. Med. 193, 105383. https://doi.org/10.1016/j.prevetmed.2021.105383

Bibliographic references

1.           Whay, H. R. & Shearer, J. K. The Impact of Lameness on Welfare of the Dairy Cow. Vet. Clin. North Am. - Food Anim. Pract. 33, 153–164 (2017).

2.           Willshire, J. A. & Bell, N. J. An economic review of cattle lameness. in Proceedings of the British Cattle Veterinary Association Congress 136–141 (British Cattle Veterinary Association, 2009).

3.           Cutler, J. H. H. et al. Producer estimates of prevalence and perceived importance of lameness in dairy herds with tiestalls, freestalls, and automated milking systems. J. Dairy Sci. 100, 9871–9880 (2017).

4.           O’Leary, N. W., Byrne, D. T., O’Connor, A. H. & Shalloo, L. Invited review: Cattle lameness detection with accelerometers. J. Dairy Sci. (2020) doi:10.3168/jds.2019-17123.

5.           Navarro, G., Green, L. E. & Tadich, N. Effect of lameness and lesion specific causes of lameness on time budgets of dairy cows at pasture and when housed. Vet. J. 197, 788–793 (2013).

Device used to identify altered behaviour and movement in lame cows on pasture.
This device combines a 3D accelerometer (59.5 Hz) and a GPS (1 Hz) attached to the cow's collar.
The 3 axes of the accelerometer are shown.