Intelligent Signal Analysis Using Case-Based Reasoning for Decision Support in Stress Management - Computational Intelligence in Healthcare: Advanced Methodologies - page 163

The step 'Verbal-stress' is initiated by asking the subject to tell about some stressful
event they experienced in life. During the second half of this step the subject thinks
about some negative stressful events in their life. In the 'Relax step', the subject is
instructed to think of something positive, either a moment in life when they were very
happy or a future event they look forward to experiencing. The 'Math-stress' step
tests the subject's reaction to directly induced stress by the clinician where the subject
is requested to count backwards. Finally, the 'relaxation step' tests if and how quickly
the subject recovers from stress. An example of the finger temperature measurement
during the six different steps in the calibration phase is shown in Fig. 11.

DeepBreath Relax

Baseline VerbalStress MathStress Relax

Time in minutes

Fig. 11. An example of the finger temperature measurement during the six different steps in the
calibration phase

The proposed CBR system works in several steps to diagnose individual sensitivity
to stress as shown in Fig. 12.

Case-Based
classification
scheme

FT sensor signal
measurement

Signal
processing to
identify
features to
formulate case

Classified FT
measurements

Fig. 12. Schematic diagram of the steps in stress diagnosis

The CBR system takes the finger temperature measurement as an input. Then,
from this sensor signal, the system identifies the essential features and formulates a
new problem case with the extracted features in a vector [71]. This new problem case
is then fed into the CBR cycle to retrieve the most similar cases. The new case
(feature vector extracted from the FT signal) in this system is then matched using