Validation of a non-invasive thermographic method for determining body temperature of rabbits in tropical production systems

Abstract

The objective was to validate thermography for determining body surface temperature (BST) in rabbits in a lowland tropical production system. Using 30 individuals, 300 measurements of inner (I) and outer (E) temperatures were taken from the right ear, nose (N), and eye (E) using a FLIR TG267 thermal imaging camera (CT) and a NJTY T600 digital infrared thermometer (TD) over 10 consecutive days at 10:00 a.m. The data were compared with rectal temperature (RT) obtained with a digital thermometer. A fixed-effects GLM was used to understand the simple effect of physiological status and measuring equipment, and a mixed-effects GLM was used, considering animal and repetition as random factors. A linear regression and a concordance matrix (Bland-Altman) were constructed assuming that the individual's RT was the body's RT. The average RT was 38.9±0.53ºC. In all cases, differences (p<0.05) were found between the temperature measured by CT (OI: 32.2±2.3ºC, OE: 30.8±2.2ºC, N: 30.0±2.1ºC and O: 31.8±1.9ºC) and TD (OI: 29.8±2.2ºC, OE: 28.8±2.1ºC, N: 28.9±1.9ºC and O: 29.6±1.8ºC) for the body regions evaluated. Mixed effects modeling showed significant differences between assessment days (p>0.05), while the interaction between day*measuring equipment was not significant (p>0.05). A smaller difference was found between TR and CT measurements, especially with the OI measurement (p<0.001). A linear regression test (R2 = 0.11) proposed the regression equation y=36.30+0.0794x, where y is the rectal temperature and x is the OI CT temperature. The results obtained support the use of infrared thermography in the inner ear region as a non-invasive method with the potential to estimate CST in rabbits.