Detection of hoof lesions using digital infrared thermography in dairy cows

The objective was to investigate infrared thermography (IRT) as a noninvasive diagnostic tool for early detection of foot pathologies in dairy cows. This was achieved by measuring changes in coronary band temperature before and after claw trimming in response to visual detection of abnormalities of the hooves. We hypothesized that by focusing on the coronary band region, IRT is able to detect lesions of the hind limbs of dairy cows associated with lameness. In this study, 626 individual observations were collected from 24 cows before and after claw trimming. Infrared thermography was used to assess the surface temperature of the coronary band (CB) region and skin (S), and the temperature difference (Delta T) between CB and S of the hind limbs. The average, minimum, and maximum surface temperatures were recorded in both regions. Temperatures of CB and S and Delta T were significantly higher in cows <= 200 d in milk than in cows >200 d in milk for all healthy hooves: 31.8 +/- 2.7 versus 29.8 +/- 3.6; 28.5 +/- 2.5 versus 27.2 +/- 3.3 degrees C, and 3.31 +/- 1.7 versus 2.51 +/- 1.3 degrees C, respectively. Temperatures of CB and S regions were positively correlated with ambient temperature. This association was best described by a linear model (R-2 = 0.92 and 0.99, respectively). The temperatures of CB and S regions were 30.3 +/- 3.2 degrees C and 27.3 +/- 2.9 degrees C; 32.1 +/- 1.7 C and 28.6 +/- 2.1 degrees C; and 33.8 +/- 1.3 degrees C and 29.9 +/- 1.8 degrees C for parlor temperatures of 12.2, 15.7, and 20.3 degrees C, respectively. In the pre- and post-trimming data analysis, a significant difference was found in temperature of the coronary band between cows with lesions and cows without lesions. A threshold value was established to determine the temperature difference between lesion and nonlesion hind claws on CB at 0.64 and 1.09 degrees C before and after claw trimming (sensitivity = 85.7%, specificity = 55.9%; and sensitivity = 80.0%, specificity = 82.9%, respectively) with the aim of detecting hoof lesions. In conclusion, the results demonstrate an increase in surface temperature of the lame limb when a hoof has a lesion.