Use of Infrared Thermography for Early Detection of Disease

These findings are an important first step in the development of a cost-effective system for detecting diseases such as Actinobacillus pleuropneumoniae, according to R. Friendship, Z. Poljak, and K. McIntosh of Ontario Veterinary College. They presented their paper at the 2009 Centralia Swine Research Update.
calendar icon 2 June 2009
clock icon 5 minute read


In a finishing barn situation, pathogens that lead to mortality can be devastating to a producer. Infectious diseases such as Actinobacillus pleuropneumoniae (APP) can display few clinical signs before pig mortality occurs. This is complicated by the economic infeasibility of inspecting individual pigs for early signs of clinical disease in a large finishing operation. A pen-level method for detecting early stages of the disease would therefore be beneficial from both a production and a herd health point of view. The objective of this study was to evaluate the use of infrared thermography as an early detection method for mortality-causing disease in a finisher barn setting.

Materials and Methods

Over a consecutive 19-day period in June 2008, 56 finisher pens in a finishing barn experiencing an outbreak of mortality due to APP were imaged using a Fluke Ti45 Infrared Camera. The finisher barn operated as a naturally ventilated, continuous flow operation divided into four rooms with a common airflow, with each pen housing 20 pigs on average. Each morning, two consecutive thermal images were taken of each pen. Humidity, environmental temperature, age class, maximum temperature within the pen, and number of mortalities for each pen were recorded. For the purposes of classification, a case was defined as a pen that had at least one mortality recorded that day. Each case pen was randomly matched with three control pens. Matching was performed with a statistical software package, using day of the study and approximate age category. The data was imported into the STATA Data analysis package and was analysed using a conditional logistic regression model. The pen-level temperature data was compared using values for the day of, day before, and two days before mortality occurred. Statistical repeatability studies were also performed on the camera to evaluate the precision of the camera upon itself.


Using the odds-ratio as a measure of association, it was found that case pens had significantly higher maximum temperature than control pens, not only on the day that the mortality was recorded (OR=3.65, P<0.0001), but also the day before (OR=3.58, P<0.0001) and two days before (OR=3.64 P=0.004) mortality (n=27). Thus, with each one degree Centigrade increase in maximum temperature of a pen, the odds of a pen having mortality increased 3.6 times. It was also found that the odds ratio was similar (OR=1.85 for day of mortality, OR=2.31 for the day before, and OR=2.45 for two days before mortality) when only considering the first time that the pen experienced mortality, and although the results were not significant, the low sample size (n=10) may have led to the non-significant P-values.

Significance to Practitioners and Producers

These findings are an important first step in the development of a cost-effective system for detecting diseases such as APP. The data collected by a producer could allow them to avoid having to institute barn-wide antibiotic protocols when an outbreak of APP occurs. Instead, a producer could isolate and treat specific pens, thereby saving costs and also preserving niche-market operations, such as those that operate antibiotic-free. As well, the operation of the thermal camera, if properly cared for, is fairly straightforward, not unlike a digital camera. It is conceivable that a producer could image pigs as a supplement to a daily routine such as feeding, while not adding a significant amount of extra time into an already busy schedule. Mounting cameras on in a barn and taking still shots over a set period of time could also be possible, thereby reducing the input required by the producer further.

For practitioners, thermal imaging provides another tool to diagnose a disease that is particularly difficult to diagnose ante-mortem. The infrared image (Figure 1, and optional digital picture), along with the relevant quantitative data, extracts to any personal computer using a standard camera SD card, where it can be reviewed, analysed and trends examined. As well, it provides an extra level of epidemiological data that allows swine practitioners to determine what occurred leading up to an outbreak, potentially leading to effective, targeted recommendations for control of disease. Thermal imaging can also indicate the early presence of potentially poor-performing pigs due to conditions such as ulcers and superficial or joint inflammation, allowing a practitioner to discuss these issues with a producer and develop an appropriate plan to maximise the herd’s welfare and overall health status.

Although preliminary to conclusively link the trends between pen temperature and finisher barn disease mortality, the results seen in this study are promising enough to warrant further investigation into this and other topics into which thermography can be implemented to assist producers and practitioners alike.

Figure 1. Thermal image taken on day 4 from pen 1L4, which experienced a high APP-suspect mortality rate during the study. The lone pig at the back right of the pen is reading a maximum temperature of 42.1°C. For reference, the ambient temperature was 24.8°C and the humidity was 62 per cent. On the day following this image, there were three recorded mortalities in this pen.

Further Reading

- You can view other papers presented at the Centralia Swine Research Update 2009 by clicking here.

Further Reading

- Find out more information on Actinobacillus pleuropneumoniae by clicking here.

May 2009
© 2000 - 2022 - Global Ag Media. All Rights Reserved | No part of this site may be reproduced without permission.