Temporal, spatial, inter-, and intra-cow repeatability of thermal imaging

Area of Science:

• Animal Science
• Veterinary Medicine
• Agricultural Engineering

Background:

• Infrared thermography offers a non-invasive method for assessing animal health and welfare.
• Understanding the variability in thermal imaging data is crucial for accurate interpretation in livestock management.

Purpose of the Study:

• To quantify within- and between-cow, operator, and day variances of temperature parameters from thermal images of Holstein-Friesian cows.
• To evaluate the repeatability and precision of thermal imaging for different anatomical areas (eye, hoof, udder).
• To determine the optimal number of replicate images for precise temperature measurements.

Main Methods:

• Three experiments were conducted involving single or multiple operators capturing thermal images of cows' eyes, hooves, and udders.
• Maximum, minimum, and average temperature parameters were extracted from thermal images.
• Mixed models were used to partition variance into error, cow, operator, and day components.
• Precision was defined by the 95% confidence interval range relative to a gold standard average temperature.

Main Results:

• The hoof was the most repeatable anatomical area, with cow variance accounting for 91.37–99.28%.
• Maximum temperature for eye and udder images, and average temperature for hoof images, exhibited the lowest error variance.
• No significant between-day variance was found for maximum hoof temperatures.
• High precision was achievable with at least 3 replicate images of the eye, udder, or hooves.

Conclusions:

• Infrared thermography can achieve high precision for bovine temperature measurements in agricultural settings with appropriate protocols.
• Standard operating procedures are essential to minimize operator variance when multiple individuals capture thermal images.
• Averaging multiple replicate images (eye, udder, hoof) enhances the reliability of thermal imaging data in cattle.