Prediction of the brain temperature from other body temperatures in hypothermia induced rats by using artificial neural networks

In this study, a microcontroller based temperature controlled hypothermia induction system is realized. The novelty of the experimental system is in using a thermoelectric Peltier cooler to bring the body temperature to hypothermic levels. The right and left ear, rectum, and brain temperatures of a rat are measured with this system. Then, an artificial neural network (ANN) is trained using temperature data to predict brain temperature from other body temperatures. It is shown that the system has the capability to predict the brain temperature with low error from rectum temperature measurements. The results from this empirical prototype also show that it is possible to use the system on humans to induce local hypothermia safely, where an interior temperature probe in the scalp is eliminated.