Prediction of Rectal Temperature Using Infrared Thermography in Holstein Dairy Cows

Introduction: The use of a rectal thermometer is the most common method of recording body temperature in dairy cows in field conditions. It is invasive, time-consuming, labour-demanding, a source of pathogen spread, and non-compliant with the requirements of present-day precision and welfare farming (Tao et al., 2021). The current work aimed to find the correlation between the thermographic temperature of different regions of interest (ROI) and rectal temperature (RT) in in-field Holstein cows. The secondary aim of the study was to find the ease of use (EoU) of an infrared thermographic camera at different ROI.

Materials and methods: Forty- two randomly selected Holstein dairy cows from Hartpury University Home Dairy Farm were recruited to collect RT and maximum and average infrared thermographic (IRT) temperatures of the eye region, muzzle, ear base (EB), tail base (TB), and vulva. RT was measured using a digital veterinary thermometer while the PK-80 camera (SATIR Europe (Ireland) Co. Ltd.) was employed to take thermographic temperatures of ROI. To avoid the influence of the anxious response of RT on thermographic measurements, the latter preceded the RT collection. Fridman’s ANOVA and Spearman’s correlation were used to find the significance of difference between means and to find the correlation respectively.

Results: In the comparison of means, the thermographic temperatures of all ROIs were significantly different from each other and lower than RT. Among all ROI temperatures, eye temperature was the highest, most consistent, and numerically closest to RT. Furthermore, the maximum eye temperature was more consistent than the average eye temperature. In the EoU score TB was significantly the highest but at the same time was most dispersed. None of the IRT temperatures registered a significant correlation with the RT.

Discussion: As the maximum IRT temperature of the eye region was the highest and most consistent among all the IRT temperatures it is suggested as an area for further research. Although TB offered most EoU, due to its highest dispersion it is unreliable for taking IRT readings, whereas eye temperature was second highest in EoU score, most consistent, and numerically closest to RT. Consequently, the eye region is suggested as the preferred site for recording IRT temperature in these animals. In Spearman’s correlation, no significant correlation was found between any of the thermographic temperatures and RT, and the influence of environmental factors, poor camera quality, and lack of longitudinal study design could have contributed to such a finding (Jansson et
al., 2021; Stukelj, Hajdinjak, and Pusnik, 2022). It is therefore suggested that at this stage infrared thermographic temperatures cannot be used to predict body temperature in these animals in field conditions. Future work might look to overcome the present limitations by designing a longitudinal study, switching off any fans before obtaining IRT readings, adding precision to camera position, and employing high-quality thermal cameras.

References:
Jansson, A. et al. (2021) ‘An investigation into factors influencing basal eye temperature in the domestic horse (Equus caballus) when measured using infrared thermography in field conditions’, Physiology and Behavior, 228, p. 113218. Available at: https://doi.org/10.1016/j.physbeh.2020.113218

Stukelj, M., Hajdinjak, M. and Pusnik, I. (2022) ‘Stress-free measurement of body temperature of pigs by using thermal imaging – Useful fact or wishful thinking’, Computers and Electronics in Agriculture, 193, p. 106656. Available at:
https://doi.org/10.1016/j.compag.2021.106656

Tao, W. et al. (2021) ‘Review of the internet of things communication technologies in smart agriculture and challenges’, Computers and Electronics in Agriculture, 189, p. 106352. Available at: https://doi.org/10.1016/j.compag.2021.106352