Infrared Thermography as a Diagnostic Test to Predict Heat Stress Events in Feedlot Cattle
By Ellen M Unruh, BS; Miles E. Theurer, DVM, PhD; Brad J. White, DVM, MS; Robert L. Larson, DVM, PhD; James S. Drouillard, PhD; Nora Schrag, DVM
Feedlot cattle frequently endure high environmental temperature-humidity index conditions in the summer months of many cattle feeding areas in North America. It has been shown that adequate heat abatement overnight is necessary to reduce heat stress events the following day. A noninvasive, remotely applied, practical method to identify animals that did not adequately cool overnight is needed to improve animal welfare and performance. Predicting which animals in a feedlot are predisposed to heat stress would allow precautionary measures to be set in place for calves at high risk for hyperthermal events, thereby improving animal welfare. The objective of this study was to determine if infrared thermography camera images taken in the morning hours after overnight heat abatement could be used to obtain data for use as a diagnostic test to predict afternoon heat stress events in feedlot cattle during times of elevated ambient temperatures.
Sixty crossbred beef heifers (mean body weight, 385.8 kg) were used for the study. No shade or shelter were available to the calves. A remote weather station at a nearby location recorded ambient temperatures and humidity readings every hour on study days. Using recorded ambient temperature and humidity, temperature humidity index (THI) was calculated. Profile digital thermal images of individual animals were captured during the 0600 hour and 1500 hour for ten days, over a 14-day period, when ambient temperature was forecasted to be above 29.4°C. Pant scores were assigned to individual animals during the 0600 hour and 1500 hour by an observer blinded to thermal images.
Relationship between infrared thermography data and pant scores were evaluated with artificial learning models. Afternoon infrared thermography data were related to afternoon pant scores. Morning infrared thermography images were not related to afternoon pant scores. Artificial learning model evaluation of morning infrared images were not highly accurate indicators for predicting afternoon heat stress events. The artificial learning models used resulted in different diagnostic sensitivity, specificity, overall accuracy, and kappa values, but all models resulted in overall low accuracy for predicting afternoon heat stress events. Morning weather conditions provided better information for predicting heat stress events compared to thermography images.
Infrared technology was identified as a potential means to objectively measure calves experiencing a heat stress event in a research setting. Using infrared technology as a diagnostic test was not an accurate assessment for predicting heat stress events in this study. More data needs to be collected and analyzed to determine if infrared technology could be used as a diagnostic tool.