Intelligent Signal Analysis Using Case-Based Reasoning for Decision Support in Stress Management - Computational Intelligence in Healthcare: Advanced Methodologies - page 167

The finger temperature measurement in Fig. 17 is shown for a student before his
master's thesis presentation. He explains that he was so stressed before the presenta-
tion that he could not recover from the stress in the next stages.
Time in minutes
Fig. 17. A student before the thesis presentation.
Three situations have been observed while collecting the FT measurement: a. fin-
ger temperature decreases with increasing stress which is the most common situation,
b. finger temperature increases with increasing stress i.e. paradoxical relation and c.
little or no changes i.e. the temperature remains stable when a person is experiencing
stress, this is exceptional but might happened for some people. In such cases clinical
expertise is important. In general, finger temperature decreases when a person is
stressed and increases during relaxation or in a non-stressed situation. So, the indica-
tion of a person's stress condition can be obtained from the individual temperature
variations during the six phases.
4.2 Finger Temperature Sensor Signal Processing
After analyzing a number of finger temperature signals, we found large individual
variations, but also a similarity in the pattern that temperature decreases during stress
and increases during relaxation for most people. That is an important feature that
needs to be identified by an automatic classification algorithm searching for “similar”
patients. In CBR we need to extract important features from the sensor signal. Choice
of the signal processing method depends mainly on the nature of the signal and the
specific problem to be solved [44]. The following subsections provide a summary of
the other traditional signal processing methods which have also been investigated
before applying CBR to analyze the biomedical signal i.e. Finger temperature sensor
signal.