Tel.+49 30 838 62261 Fax.+49 30 838 62512
The well-being of animals, especially those used for food, has become a focus of public attention. Early detection of sickness may lead to better therapeutic results and, consequently, to increased animal welfare. Furthermore, because of increasing herd sizes and decreasing staff, individual animal care is becoming less common. Therefore, continuous monitoring of animal behaviour seems to become more and more important to improve prognoses. Following, early treatment of disease can be performed. One meaningful parameter for monitoring the health of the animals is body temperature. Rectal temperature measurements are invasive and time consuming; therefore, new methods would help improve the automatic detection of the body temperature.
The overall objective of this dissertation was to validate various temperature devices to continuously monitor the body temperature of individual sows in a less invasive manner and to compare these data to rectal temperature data. Furthermore, specific behavioural patterns during fever were monitored to determine which changes in behaviour occur during disease states. These patterns could be used as a basis for automatic health detection systems.
An infrared camera and an infrared thermometer were used to measure body surface temperature at various body locations. The eye and the back of the ear were determined to be good locations to measure the body surface temperature of sows in a contactless, noninvasive and automatic manner. The median differences between the infrared camera and the rectal temperatures were lower than those between the infrared thermometer temperatures and the rectal temperatures. The sensitivity of the measurements with the infrared camera was greater when measured at the back of the ear, followed by the eye, when compared to the other locations. Next to the eye and the back of the ear, the mammary gland and the vulva also generated acceptable results; however, these locations are not as practical. Unfortunately, this study showed that single body temperature measurements using the infrared technique are not able to detect fever in pigs. This work also presented the limits of this technique. One of the major difficulties is to aim the infrared device at the right measuring spot. Regardless, when measuring with just an infrared thermometer, fixed or unfixed, the appropriate spot and diameter at which to aim cannot be secured. However, an infrared camera has better aim, and infrared movies provide more information. Regardless of the limitations, these methods provide an advantage and allow for contactless continuous temperature measurements in animals.
Furthermore, to test alternative methods for automatic measurement of body temperature, sows were vaccinated with a commercial vaccine to induce fever and mild sickness behaviours. It was shown that the rhinitis atrophicans vaccine was a low risk method for immune stimulation and for the induction of mild sickness behaviours and fever in gilts. During this trial, a vaginal logger was evaluated. A positive correlation between rectal temperatures and the temperatures from the logger were detected in the 21 vaccinated gilts. Hence, the logger might be a useful and less invasive tool for core body temperature monitoring. Unfortunately, due to urination and defecation, the measuring device often slipped out and needed to be reinserted. Therefore, this device might be useful in research, but not for commercial use.
While under the influence of the vaccine, infrared thermometers, also known as pyrometers, were also tested. The pyrometer was aimed at the orbital area of each gilt and was able to collect continuous and contactless measurements. The pyrometer temperatures were taken while the gilts were drinking. Correlation was observed when continuous temperature monitoring was performed in two hour intervals. However, no satisfactory results could be retrieved. It was shown that the standard error of the measurements with both devices (the infrared thermometer and the vaginal logger) increased during the night and decreased during the day. This variation is likely linked to drinking frequency, as the sows clearly demonstrated a rhythmic pattern of drinking throughout the day and night. A short increase in body surface temperature could not be automatically detected using the pyrometer. The variability compared to the vaginal logger was too high; however, this variability might be different for longer-term temperature increases. However, using the rhinitis atrophicans vaccine to induce mild sickness behaviour and an increase in gilts body temperature is less risky than the injection of LPS. Using this vaccine is a good alternative method of inducing fever without causing severe side effects so as to monitor sickness behaviour or to evaluate sensors.
As an alternative method for continuous temperature measurements, the vaginal logger may be a good choice, when its limits are taken into account. These limits include the fixation of the logging device and the loss of the logger during urination and defecation. Therefore, this device could be used for research endeavours, but is not suitable for use in practical application. The detection of illness behaviours showed that parameters such as lying down and water intake changed while the animals exhibited higher temperatures. For the detection of these parameters, long-term measurements of body surface temperature at defined locations using an infrared camera could provide helpful information for the early diagnosis of disease. Thus, early treatment could be performed, leading to better performance of the animals and the improved well being of the sow.